feat(agents): add Task Challenger adversarial questioning agent

[rezatnoMsirhC]

feat(agents): add Task Challenger adversarial questioning agent

Description

Added Task Challenger — an adversarial questioning agent that reads .copilot-tracking/ artifacts cold and interrogates every decision, boundary, and assumption through structured What/Why/How questions. The agent does not validate, suggest, coach, or guide; it asks.

The agent operates in four phases. In Phase 1 (Scope), it discovers what to challenge through a five-level cascade: existing .copilot-tracking/ artifacts, pr-reference.xml, git branch history, domain-based workspace search, and finally direct user input. Terminal access is limited to this phase only. Phase 2 (Read Artifacts) silently reads plans, changes, research, and reviews from .copilot-tracking/. Phase 3 (Identify Challenge Areas) silently selects the 5–7 areas with the highest density of unexamined assumptions — this list is never disclosed to the user. Phase 4 (Challenge) issues exactly one question per response using the structure [What/Why/How] + [noun subject] + [verb] + [open object]?, probes each answer up to twice before marking a point unresolved, and handles skip signals ("Go next", "Skip", etc.) without acknowledgment.

A Challenge Tracking Document is created at Phase 4 entry under .copilot-tracking/challenges/{{YYYY-MM-DD}}/{{topic}}-challenge.md. It captures metadata, confirmed scope, identified challenge areas, a Q&A log with verbatim answers, probe exchanges, and an Unresolved Items table.

The companion task-challenge.prompt.md provides four optional inputs (plan, changes, research, focus) that pre-position scope at invocation; the agent falls back to artifact discovery when none are supplied. Downstream integration with Task Researcher (challenges-directory lookup as primary research scope) and Task Reviewer (🥊 Challenge handoff) is planned for a follow-up issue.

Both hve-core and hve-core-all collections and plugins were updated in lockstep, and .github/copilot-instructions.md was updated to register .copilot-tracking/challenges/ as a known tracking directory.

Related Issue(s)

Closes #1212

Type of Change

Select all that apply:

Code & Documentation:

• Bug fix (non-breaking change fixing an issue)
• New feature (non-breaking change adding functionality)
• Breaking change (fix or feature causing existing functionality to change)
• Documentation update

Infrastructure & Configuration:

• GitHub Actions workflow
• Linting configuration (markdown, PowerShell, etc.)
• Security configuration
• DevContainer configuration
• Dependency update

AI Artifacts:

• Reviewed contribution with prompt-builder agent and addressed all feedback
• Copilot instructions (.github/instructions/*.instructions.md)
• Copilot prompt (.github/prompts/*.prompt.md)
• Copilot agent (.github/agents/*.agent.md)
• Copilot skill (.github/skills/*/SKILL.md)

Note for AI Artifact Contributors:

• Agents: Research, indexing/referencing other project (using standard VS Code GitHub Copilot/MCP tools), planning, and general implementation agents likely already exist. Review .github/agents/ before creating new ones.
• Skills: Must include both bash and PowerShell scripts. See Skills.
• Model Versions: Only contributions targeting the latest Anthropic and OpenAI models will be accepted. Older model versions (e.g., GPT-3.5, Claude 3) will be rejected.
• See Agents Not Accepted and Model Version Requirements.

Other:

• Script/automation (.ps1, .sh, .py)
• Other (please describe):

Sample Prompts (for AI Artifact Contributions)

User Request:

"Challenge this implementation" with the Task Challenger agent selected, or invoke /task-challenge directly. Optional arguments: /task-challenge plan=.copilot-tracking/plans/my-plan.md focus=authentication.

Execution Flow:

1. Phase 1 (Scope): Discovers what to challenge through a five-level cascade — .copilot-tracking/ tracking artifacts, pr-reference.xml, git log/diff/status (via git branch --show-current, git log ..HEAD --oneline, git diff --stat), domain-based repo search, or direct user prompt. Presents a factual scope summary and waits for explicit user confirmation before proceeding. Terminal commands are run only during this phase.
2. Phase 2 (Read Artifacts): Silently reads .copilot-tracking/plans/, changes/, research/, and reviews/.
3. Phase 3 (Identify Challenge Areas): Silently identifies 5–7 assumption-dense areas. List is never disclosed to the user.
4. Phase 4 (Challenge): Creates .copilot-tracking/challenges/{{YYYY-MM-DD}}/{{topic}}-challenge.md. Issues one [What/Why/How] + […]? question per response. Probes each answer up to twice; marks unresolved after two probes with no new depth. Advances silently on skip signals. Updates the tracking document throughout the session.
5. On completion, the Compact handoff summarizes state (including complete Q&A and unresolved items) and defaults the next step to Task Researcher.

Output Artifacts:

# Challenge Session: task-challenger-agent

**Date**: 2026-04-07
**Scope source**: Level 1 — .copilot-tracking/ artifacts
**Related artifacts**: .copilot-tracking/plans/..., .copilot-tracking/changes/...

## Confirmed Scope
...

## Q&A Log
### Area: scope boundaries
**Q**: What does the five-level cascade exclude?
**A**: [verbatim answer]
  **Probe**: How is an empty artifacts folder distinguished from a missing one?
  **A**: [verbatim answer]

## Unresolved Items

| # | Area             | Last Question Asked                                        | Reason                        |
|---|------------------|------------------------------------------------------------|-------------------------------|
| 1 | scope boundaries | How is an empty folder distinguished from a missing one?   | No new depth after two probes |

Success Indicators:

• Agent presents a factual scope summary and waits for explicit confirmation before entering Phase 4.
• Each Phase 4 response contains exactly one question with no preamble, no affirmation, and no suggestion.
• The challenge tracking document is created at Phase 4 entry and updated after each Q&A exchange.

For detailed contribution requirements, see:

• Common Standards: docs/contributing/ai-artifacts-common.md - Shared standards for XML blocks, markdown quality, RFC 2119, validation, and testing
• Agents: docs/contributing/custom-agents.md - Agent configurations with tools and behavior patterns
• Prompts: docs/contributing/prompts.md - Workflow-specific guidance with template variables
• Instructions: docs/contributing/instructions.md - Technology-specific standards with glob patterns
• Skills: docs/contributing/skills.md - Task execution utilities with cross-platform scripts

Testing

All automated checks were run during PR generation.

Check	Command	Result
Markdown linting	npm run lint:md	✅ Passed — 0 errors across 196 files
Spell checking	npm run spell-check	✅ Passed — 0 issues across 298 files
Frontmatter validation	npm run lint:frontmatter	✅ Passed — 0 errors, 0 warnings across 490 files
Skill structure validation	npm run validate:skills	✅ Passed — 14 skills, 0 errors
Link validation	npm run lint:md-links	❌ Pre-existing SECURITY.md failure on main — not introduced by this PR
PowerShell analysis	npm run lint:ps	✅ Passed — all files clean
Plugin freshness	npm run plugin:generate	✅ Passed — two table-formatting fixups applied (see Additional Notes)
Docusaurus tests	npm run docs:test	⏭️ Skipped — jest not installed; pre-existing environment limitation

Security analysis:

• Terminal access is explicitly contained to Phase 1 only; the agent declares execute/runInTerminal and execute/getTerminalOutput in tools but the agent body restricts their use to the Scope phase.
• No write or edit tools declared — the agent cannot modify files.
• Challenge tracking documents write to .copilot-tracking/challenges/, which is gitignored per existing repo convention.
• No new npm, pip, or external package dependencies.
• No credentials, tokens, or sensitive data introduced.

Sample Run

Scope Confirmation Phase

Conversational Q&A Phase

Checklist

Required Checks

• Documentation is updated (if applicable) — docs/agents/README.md not yet updated to include Task Challenger; deferred to follow-up issue
• Files follow existing naming conventions
• Changes are backwards compatible (if applicable)
• Tests added for new functionality (if applicable) (N/A — no test infrastructure for agent/prompt files)

AI Artifact Contributions

• Used /prompt-analyze to review contribution
• Addressed all feedback from prompt-builder review
• Verified contribution follows common standards and type-specific requirements

Required Automated Checks

The following validation commands must pass before merging:

• Markdown linting: npm run lint:md
• Spell checking: npm run spell-check
• Frontmatter validation: npm run lint:frontmatter
• Skill structure validation: npm run validate:skills
• Link validation: npm run lint:md-links (pre-existing SECURITY.md failure on main)
• PowerShell analysis: npm run lint:ps
• Plugin freshness: npm run plugin:generate
• Docusaurus tests: npm run docs:test (jest not installed; pre-existing environment limitation)

Security Considerations

• This PR does not contain any sensitive or NDA information
• Any new dependencies have been reviewed for security issues (N/A — no new dependencies)
• Security-related scripts follow the principle of least privilege (N/A — no security scripts modified)

Additional Notes

• plugin:generate applied two minor fixups during validation: markdown table column padding in task-challenger.agent.md and a missing task-challenge row in plugins/hve-core-all/README.md. Both are present as unstaged local modifications and should be committed before merging.
• docs/agents/README.md names five RPI agents (task-researcher, task-planner, task-implementor, task-reviewer, and the RPI orchestrator) but does not yet mention Task Challenger. A follow-up issue will track this documentation update.
• The lint:md-links failure on SECURITY.md is pre-existing on main and unrelated to this PR.

[codecov-commenter]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 87.65%. Comparing base (3aded28) to head (6416f3d).

Additional details and impacted files

@@            Coverage Diff             @@
##             main    #1315      +/-   ##
==========================================
- Coverage   87.66%   87.65%   -0.02%     
==========================================
  Files          61       61              
  Lines        9328     9328              
==========================================
- Hits         8177     8176       -1     
- Misses       1151     1152       +1     

Flag	Coverage Δ
pester	85.22% <ø> (-0.02%)	⬇️

Flags with carried forward coverage won't be shown. Click here to find out more.
see 1 file with indirect coverage changes

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.

[github-actions]

PR Review: feat(agents): add Task Challenger adversarial questioning agent

This is a well-conceived and thoroughly described feature. The adversarial questioning model, four-phase protocol, and data contract from challenger outputs to researcher inputs are all clearly specified. The automated checks pass. Two issues need to be resolved before merging.

---

Issue Alignment

The PR description uses Related to #1212 rather than Fixes #1212 or Closes #1212. The PR template instructs contributors to use Fixes # or Closes # syntax. If this PR fully delivers the feature described in the linked issue, consider updating to Closes #1212 so the issue is automatically closed on merge. If the issue is intentionally left open (e.g., it tracks ongoing work), a brief note explaining why would clarify intent.

---

PR Template Compliance

⚠️ AI Artifact Contributions checklist — all three items unchecked

The PR adds AI artifact files (a new agent and a new prompt), which requires completing the AI Artifact Contributions checklist:

* [ ] Used `/prompt-analyze` to review contribution
* [ ] Addressed all feedback from `prompt-builder` review
* [ ] Verified contribution follows common standards and type-specific requirements

All three remain unchecked. These checkboxes represent a required quality gate for AI artifact contributions. Please complete the /prompt-analyze review, address any feedback, and check these items before requesting re-review.

ℹ️ Documentation checkbox

The Documentation is updated (if applicable) checkbox is unchecked. The Additional Notes section acknowledges that docs/agents/README.md currently omits the Task Challenger. While deferring the docs update to a follow-up is a reasonable call, the checkbox should carry an inline (N/A — deferred to follow-up issue) annotation to make that intent explicit, per the checklist conventions used elsewhere in the template.

---

Coding Standards

❌ disable-model-invocation: true missing from task-challenger.agent.md frontmatter (inline comment on line 4)

The agent declares execute/runInTerminal and execute/getTerminalOutput in tools:, writes files to .copilot-tracking/challenges/, and runs git commands during Phase 1. This makes it a side-effecting agent. task-reviewer.agent.md — a direct peer — sets disable-model-invocation: true for the same reason. The prompt-builder instructions require this field for side-effecting and explicitly-invoked agents. Add disable-model-invocation: true to the frontmatter.

---

Code Quality

💡 Duplicate "Response Format" section (inline comment on line ~214)

The file defines the Phase 4 response format in two places: a nested #### Response Format inside ### Phase 4: Challenge, and a top-level ## Response Format at the end of the file. Both sections state the same one-question rule with the same examples. Duplicate instructions create a maintenance hazard — a future edit to one copy may silently diverge from the other. Consolidate into the nested section, which is already the more contextually natural location.

---

Action Items

1. Required — Check all three AI Artifact Contributions checklist items after completing the /prompt-analyze review.
2. Required — Add disable-model-invocation: true to the task-challenger.agent.md frontmatter.
3. Suggested — Remove the duplicate top-level ## Response Format section; retain only the nested #### Response Format under Phase 4.
4. Minor — Annotate the Documentation checkbox with (N/A — deferred to follow-up) to make the deferral explicit.

Note

🔒 Integrity filter blocked 1 item

The following item were blocked because they don't meet the GitHub integrity level.

• #1212 issue_read: has lower integrity than agent requires. The agent cannot read data with integrity below "approved".

To allow these resources, lower min-integrity in your GitHub frontmatter:

tools:
  github:
    min-integrity: approved  # merged | approved | unapproved | none

Generated by PR Review for issue #1315

[katriendg]

Thanks @rezatnoMsirhC for this new challenger agent proposal. I like many of the ideas, but I do feel it is adding some broad changes to RPI and the related agents, which we may want to be careful about, in the sense that it needs to be very clear and straightforward to users.

I left a few notes as I first reviewed it, probably would want to use it in action before we finalize in the coming days.

At first I would only release in Experimental mode to ensure we get enough user testing and feedback.

If this agent becomes a more integrated part of RPI, we need to ensure docs are updated to correctly explain how to use this agent, and how it should especially be used with the RPI workflow. Again if experimental we often allow for agents without full docs, but I would not release it without documentation.

Could you also run /prompt-analyze and attach your new agent and prompt files? I see the checkbox is not checked in the PR so I imagine you may not have run it yet. It may also give you some recommendations.

[rezatnoMsirhC]

Thanks @rezatnoMsirhC for this new challenger agent proposal. I like many of the ideas, but I do feel it is adding some broad changes to RPI and the related agents, which we may want to be careful about, in the sense that it needs to be very clear and straightforward to users.

I left a few notes as I first reviewed it, probably would want to use it in action before we finalize in the coming days.

At first I would only release in Experimental mode to ensure we get enough user testing and feedback.

If this agent becomes a more integrated part of RPI, we need to ensure docs are updated to correctly explain how to use this agent, and how it should especially be used with the RPI workflow. Again if experimental we often allow for agents without full docs, but I would not release it without documentation.

Could you also run /prompt-analyze and attach your new agent and prompt files? I see the checkbox is not checked in the PR so I imagine you may not have run it yet. It may also give you some recommendations.

Thank you for your extensive review! This is my first agent, so I'm still learning a lot.

I've run /prompt-analyze now and I've fixed the issues.

[katriendg]

Thanks for your changes, and moving the artifacts into the hve-core collection.

Please note we need to keep this in experimental maturity, so before we merge that key needs to be added in the **.collection.yml files. I left comments inline for this.

Please also update the .github/CUSTOM-AGENTS.md file with this new agent documentation.

For the rest it looks like this really interesting addition, looking forward to testing it!

[github-actions]

PR Review: feat(agents): add Task Challenger adversarial questioning agent

This PR introduces an interesting and well-conceived adversarial questioning agent. The prompt design, question framework, prohibited behaviors list, and phase structure are thorough and clearly authored. The collection/plugin lockstep updates are correct. Two blocking issues and two minor issues require attention before this is merge-ready.

---

Issue Alignment

The PR links to #1212 using "Related to #1212" rather than "Fixes #1212" or "Closes #1212". The PR template explicitly instructs linking issues with Fixes # or Closes # to trigger auto-close on merge. If this PR is meant to resolve #1212, please update the reference. If the issue intentionally stays open (e.g., it tracks a broader epic), the current phrasing is acceptable but should be clarified.

---

PR Template Compliance

The template is well-filled overall, but the Description section contains inaccurate claims about files that were not changed in this PR:

"Task Researcher was updated to check .copilot-tracking/challenges/ at the start of Phase 1 and, when a challenge document is present, treat its Q&A log and unresolved items as the primary research scope — establishing a formal data contract from challenger outputs to researcher inputs."

"Task Reviewer received a new 🥊 Challenge handoff that routes to Task Challenger via /task-challenge."

Neither task-researcher.agent.md nor task-reviewer.agent.md is in the changed file list. Inspection of both files confirms neither contains the described changes. The "data contract" and the Task Reviewer handoff are described as shipped features but were not implemented.

This is a blocking issue: the PR description advertises capabilities that do not exist in the diff. Either:

1. The updates to task-researcher and task-reviewer must be added to this PR, or
2. The description must be corrected to remove those claims and note them as planned follow-up work.

---

Coding Standards Review

The agent file follows the prompt-builder.instructions.md authoring conventions well: frontmatter fields are present, the phase/step structure is clearly delineated, prohibited behaviors are explicit, and the question framework is consistently formatted.

One objective violation is flagged as a blocking issue — see inline comments.

---

Code Quality Findings

🔴 Blocking: Tool list vs. file creation contradiction

The declared tool list (read, search, execute/runInTerminal, execute/getTerminalOutput) does not include any write or create-file capability. Per prompt-builder.instructions.md, when tools: is specified, only listed tools are available. Yet the agent body requires creating and continuously updating a challenge tracking document:

• Core Principles (line 38): "Always create the challenge tracking document... it is always created, not optional. Update it throughout the session."
• Phase 4 Protocol (line 206): "At Phase 4 entry, create the challenge tracking document..."

The PR description frames the missing write tools as a security positive — but then describes a mandatory tracking document. These two positions are mutually exclusive. This must be resolved before the agent can function as documented. See inline comments on lines 5 and 206.

🟡 Minor: Unstaged plugin fixups must be committed before merge

The Additional Notes section correctly identifies that plugin:generate applied two fixups (table column padding in task-challenger.agent.md, missing row in plugins/hve-core-all/README.md) that remain as unstaged local modifications. These must be committed into this branch before merging.

---

Action Items

1. Resolve the tools/write contradiction — either add write tools and update the security analysis, or remove the tracking document requirement and update the description to match.
2. Correct the PR description — either implement the task-researcher and task-reviewer updates, or remove those claims and mark them as follow-up work.
3. Commit the plugin fixups before merging.
4. (Optional) Update docs/agents/README.md to mention Task Challenger, as noted in Additional Notes.

Note

🔒 Integrity filter blocked 1 item

The following item were blocked because they don't meet the GitHub integrity level.

• #1212 issue_read: has lower integrity than agent requires. The agent cannot read data with integrity below "approved".

To allow these resources, lower min-integrity in your GitHub frontmatter:

tools:
  github:
    min-integrity: approved  # merged | approved | unapproved | none

Generated by PR Review for issue #1315 · ● 4.1M

[github-actions]

⚠️ Tool list is missing write capabilities, but Phase 4 requires file creation

The declared tool set [read, search, execute/runInTerminal, execute/getTerminalOutput] does not include any write, create_file, or equivalent capability. However, the agent body mandates file creation in two places:

• Core Principles (line 38): "Always create the challenge tracking document... This document is the session record — it is always created, not optional. Update it throughout the session."
• Phase 4 Protocol (line 206): "At Phase 4 entry, create the challenge tracking document at .copilot-tracking/challenges/..."

The PR description explicitly frames the missing write tools as a security positive ("the agent cannot modify files"), which directly contradicts the requirement to create and continuously update a tracking document.

One of these must be resolved:

1. Add write/create tools if the tracking document is a required feature, and update the security analysis accordingly.
2. Remove the tracking document requirement from Core Principles and Phase 4 if the write restriction is truly intentional, and update the PR description to acknowledge that no session record is persisted.

As written, the agent will either fail silently when it attempts to create the document, or the tool restriction is non-enforceable — neither is acceptable.

[github-actions]

⚠️ File creation instruction contradicts the declared tool list

This line instructs the agent to create the challenge tracking document, but the frontmatter tool list (line 5) does not include any write or create-file tool. Since tools: in an agent frontmatter restricts available capabilities to only the listed entries, the model will have no mechanism to fulfill this instruction.

This line, along with the "Update it throughout the session" instruction on line 208, must be reconciled with the tool list before this agent can function as documented. See the comment on line 5 for the two resolution paths.

[github-actions]

Review: feat(agents): add Task Challenger adversarial questioning agent

The core Task Challenger agent is well-designed. The behavioral contracts are clearly stated, the prohibited-behavior list is specific and enforceable, the question framework is coherent, and the four-phase protocol is internally consistent. The collection and plugin updates are in order. However, two blocking issues must be resolved before this PR is ready to merge.

---

❌ Issue Alignment

The PR is linked as "Related to #1212" rather than "Fixes #1212" or "Closes #1212" as the PR template requests. If this is intentional (the issue is not fully resolved by this PR), that is acceptable — but please update the template comment to reflect why a closing keyword was not used, so reviewers understand the intent.

---

❌ PR Template Compliance — Incorrect Description of Changes (Blocking)

The PR description states:

Task Researcher was updated to check .copilot-tracking/challenges/ at the start of Phase 1 and, when a challenge document is present, treat its Q&A log and unresolved items as the primary research scope.

Task Reviewer received a new 🥊 Challenge handoff that routes to Task Challenger via /task-challenge.

Neither of these changes is present in the diff. Confirmed by comparing file SHAs on the PR head branch against main:

• .github/agents/hve-core/task-reviewer.agent.md — identical SHA on both branches; no 🥊 Challenge handoff added.
• .github/agents/hve-core/task-researcher.agent.md — identical SHA on both branches; no challenges-directory check added.

The PR description must accurately reflect the actual changes. Either:

1. Add the missing changes — update task-reviewer.agent.md with the 🥊 Challenge handoff and task-researcher.agent.md with the challenges-directory lookup, then update the collection YAMLs and run npm run plugin:generate again, OR
2. Correct the description — remove those claims from the PR body and clearly scope this PR as delivering the agent only, with a follow-up issue for the downstream integrations.

Note that without the Task Reviewer handoff, users have no discoverable navigation path from the review workflow to the challenger. The described data contract (challenger → researcher) is also incomplete without the Task Researcher update.

---

❌ PR Template Compliance — Documentation Checkbox Contradicts Acknowledged Gap (Blocking)

The Checklist section has [x] Documentation is updated (if applicable) checked, but the Additional Notes section explicitly acknowledges:

docs/agents/README.md names five RPI agents ... but does not yet mention Task Challenger. Consider a follow-up to update that reference.

Documentation is applicable here. Either:

1. Update docs/agents/README.md in this PR to include Task Challenger, OR
2. Uncheck the documentation box and open a follow-up issue, referencing it in the PR description.

Leaving a checked checkbox that contradicts an acknowledged gap misleads reviewers about the completeness of the contribution.

---

⚠️ Coding Standards Observations

Two inline comments were left on the changed files:

• task-challenger.agent.md line 5 — execute/runInTerminal and execute/getTerminalOutput are declared in frontmatter without a platform-level scope restriction. The Phase 1-only constraint lives entirely in LLM-interpreted instructions. Consider adding an explicit in-agent guard or a prominent Core Principles statement to reduce the risk of behavioral drift in long sessions.
• task-challenge.prompt.md line 20 — Requirement 3 duplicates behavior already defined in the agent's Phase 4 completion summary. Per prompt authoring conventions, Requirements sections should extend rather than restate the parent agent's protocol.

---

✅ What Looks Good

• Agent frontmatter is well-formed: name, description with attribution, disable-model-invocation: true, tools, handoffs all present and correctly structured.
• The Prohibited Behaviors list is precise and operationally clear.
• The Probing Strategy (two-probe cap before marking unresolved) is a clean, well-scoped behavioral contract.
• Phase 1 scope-discovery cascade (5 levels with explicit fallback chain) is thorough.
• Challenge tracking document schema is well-defined with all required sections.
• maturity: experimental correctly set in both collection YAMLs.
• <!-- markdownlint-disable-file --> correctly specified for tracking document output.
• npm run plugin:generate was run and plugin outputs are consistent.
• All automated checks passed (markdown lint, spell check, frontmatter, PowerShell, skills validation).

---

Action Items

1. Resolve the description/diff discrepancy for task-reviewer.agent.md and task-researcher.agent.md (add changes or correct the description).
2. Resolve the documentation checkbox contradiction (update docs/agents/README.md or uncheck the box and open a follow-up issue).
3. Address the inline comments on terminal tool scope and duplicate prompt requirement.

Note

🔒 Integrity filter blocked 1 item

The following item were blocked because they don't meet the GitHub integrity level.

• #1212 issue_read: has lower integrity than agent requires. The agent cannot read data with integrity below "approved".

To allow these resources, lower min-integrity in your GitHub frontmatter:

tools:
  github:
    min-integrity: approved  # merged | approved | unapproved | none

Generated by PR Review for issue #1315 · ● 2.6M

[github-actions]

PR Review: feat(agents): add Task Challenger adversarial questioning agent

---

Issue Alignment ✅

PR correctly targets #1212. The implementation scope matches the issue intent: an adversarial questioning agent that interrogates .copilot-tracking/ artifacts through structured What/Why/How questions without coaching, suggesting, or validating.

No scope creep detected in the 13 changed files.

---

PR Template Compliance ✅

• Description is thorough and accurately describes the change.
• Related Issue(s): Closes #1212 ✅
• Type of Change: "New feature", "Copilot prompt", "Copilot agent" all correctly checked ✅
• Sample Prompts section is complete and detailed ✅
• Testing table is present and results are explained ✅
• AI Artifact Contributions checklist is fully checked ✅
• Unchecked items (Documentation is updated, Tests added, lint:md-links, docs:test) are accompanied by explanations — acceptable for this contribution ✅

---

Coding Standards ❌

Critical: write tools missing from tools: frontmatter — see inline comment on line 5 of task-challenger.agent.md.

Per prompt-builder.instructions.md: when tools: is specified, it restricts the agent to only those tools. The declared list (read, search, execute/runInTerminal, execute/getTerminalOutput) contains no write or create capability. The agent's protocol explicitly requires creating and continuously updating .copilot-tracking/challenges/{{YYYY-MM-DD}}/{{topic}}-challenge.md throughout Phase 4. This will silently fail at runtime.

Suggested fix: add create_file (or the equivalent write tool name used in this repo) to the tools: list, or explicitly remove the tracking document creation steps from the agent protocol if write access is intentionally excluded.

---

Code Quality ⚠️

Unstaged plugin fixups must be committed before merging.

The "Additional Notes" section acknowledges: "plugin:generate applied two minor fixups — markdown table column padding in task-challenger.agent.md and a missing task-challenge row in plugins/hve-core-all/README.md — present as unstaged local modifications and should be committed before merging."

These are not in the PR commits. Per repo convention, regenerated plugin outputs must be committed as part of the PR. Please run npm run plugin:generate locally, commit the resulting changes, and push before merging.

---

Action Items

#	Severity	Item
1	🔴 Critical	Add a write/create tool to the tools: list, or reconcile the protocol so it does not require file writes
2	🟡 Required	Commit the unstaged plugin:generate fixups as part of this PR
3	⚪ Minor	Fix "no leading" → "no leading questions" in collections/hve-core.collection.md (regenerate plugins after)

Note

🔒 Integrity filter blocked 1 item

The following item were blocked because they don't meet the GitHub integrity level.

• #1212 issue_read: has lower integrity than agent requires. The agent cannot read data with integrity below "approved".

To allow these resources, lower min-integrity in your GitHub frontmatter:

tools:
  github:
    min-integrity: approved  # merged | approved | unapproved | none

Generated by PR Review for issue #1315 · ● 1.3M

[github-actions]

Automated Quality Review

Overview

The implementation is well-structured and the agent/prompt files follow the prompt-builder conventions correctly. The collection manifests, plugin outputs, and supporting documentation are all properly updated. Two PR description accuracy issues need to be addressed before this can merge.

---

✅ Issue Alignment

The PR adds a Task Challenger adversarial questioning agent that closes #1212. The implementation matches the intent described in the PR body with a clearly defined four-phase protocol, question framework, prohibited behaviors, and challenge tracking document schema.

---

❌ PR Template Compliance

Documentation checkbox is incorrectly unchecked

The "Documentation is updated (if applicable)" checklist item is unchecked with the note: "docs/agents/README.md not yet updated to include Task Challenger; deferred to follow-up issue." However, the diff shows docs/agents/README.md was updated in this PR:

• The RPI Orchestration agent count changed from 5 → 6
• task-challenger was added to the prose description of the RPI agents

Additionally, collections/hve-core.collection.md, plugins/hve-core/README.md, plugins/hve-core-all/README.md, and .github/CUSTOM-AGENTS.md were all updated. Please check the "Documentation is updated" box.

Security section contains a factual inaccuracy

The Security Considerations section asserts: "No write or edit tools declared — the agent cannot modify files." This is false. The agent frontmatter (line 5) declares edit/createFile and edit/editFiles. These tools are appropriate — the agent needs them to create .copilot-tracking/challenges/ tracking documents — but the claim in the PR description is incorrect and misleading for security reviewers who rely on these statements.

---

✅ Coding Standards

Agent and prompt files follow the prompt-builder.instructions.md conventions:

• disable-model-invocation: true is correctly set
• name: and description: frontmatter fields present with attribution suffix
• Handoff declarations reference agents by human-readable name: values (Task Researcher, Task Planner, Task Implementor)
• maturity: experimental is set in the collection manifests only (not in artifact frontmatter), per convention
• No activation line in the prompt file (correct, since it delegates to a custom agent whose protocol is self-contained)

---

✅ Code Quality

• Phase-based protocol is clearly scoped; terminal access is correctly restricted to Phase 1 only in the agent body
• Prohibited behaviors are explicit and properly scoped to the Challenge Phase
• Probe strategy (two probes before marking unresolved) is well-defined
• Five-level scope discovery cascade handles all edge cases including empty artifact directories
• Skip signal handling is specified without acknowledgment, consistent with the agent's no-preamble design

---

🔧 Required Action Items

1. Check the "Documentation is updated" checklist box — documentation was updated in this PR across docs/agents/README.md, collection markdown files, and plugin READMEs.
2. Correct the Security Considerations section — update the bullet point to accurately state that the agent declares edit/createFile and edit/editFiles tools for creating .copilot-tracking/challenges/ session artifacts.

Note

🔒 Integrity filter blocked 1 item

The following item were blocked because they don't meet the GitHub integrity level.

• #1212 issue_read: has lower integrity than agent requires. The agent cannot read data with integrity below "approved".

To allow these resources, lower min-integrity in your GitHub frontmatter:

tools:
  github:
    min-integrity: approved  # merged | approved | unapproved | none

Generated by PR Review for issue #1315 · ● 1.4M

[github-actions]

The tools declaration correctly includes edit/createFile and edit/editFiles — the agent needs these to create challenge tracking documents in .copilot-tracking/challenges/. However, the PR description's Security Considerations section states:

"No write or edit tools declared — the agent cannot modify files."

This claim is factually incorrect. Edit tools are declared here and are appropriate for the agent's purpose. Please update the security section to accurately reflect that the agent declares edit tools for creating challenge session artifacts.