[mayankb2103]

Description

Adds /recap, an explicitly invoked skill that provides a quick reminder of the current Copilot session in 1-2 short sentences, capped at 400 characters.

The recap uses only conversation context already available to the assistant. It summarizes the session goal and meaningful progress, distinguishes completed work from plans, and mentions established pending work or blockers. It handles empty or incomplete history without guessing, and does not call tools, inspect files or other sessions, modify anything, or resume unfinished work.

The PR contains only the skill and its generated entry in docs/README.skills.md. No scripts, dependencies, plugins, or hooks are added.

Validation

• npm run skill:validate - all 458 skills pass.
• npm run build - passes; generated catalog includes recap.
• git diff --check - passes.
• Tested the identical personal skill with /recap in a live Copilot CLI session: it returned two concise sentences summarizing the session, with no tool calls.
• Ran the repository line-ending script; its GNU sed -i syntax is incompatible with macOS BSD sed, so applied portable LF normalization to the two contribution files.

Pull Request Checklist

• Read and followed CONTRIBUTING.md.
• No paid services are involved.
• Adds a new skill in the correct directory with the required naming and frontmatter.
• Tested with GitHub Copilot.
• Ran the build and updated the generated skill catalog.
• Targets main.

Type of Contribution

• New skill file.

This contribution is submitted under the repository MIT License and Code of Conduct.

[github-actions]

🔒 PR Risk Scan Results

Scanned 1 changed file(s).

Severity	Count
🔴 High	0
🟠 Medium	0
ℹ️ Info	0

✅ No matching risk patterns were detected in changed files.

This is an automated soft-gate report. Findings indicate review targets and do not block merge by themselves.

[github-actions]

🚦 Submission status: 👀 Ready for review

Risk tier: merge-risk:medium — New or substantially changed resource without privileged execution
Required to merge: passing submission-gate checks plus 1 approval from reviewers with write access.

Why this tier

• Adds 1 file(s), e.g. skills/recap/SKILL.md

Automated checks

Check	Status	Details
Line endings	✅ Passed	Passed · logs
Spelling	✅ Passed	Passed · logs
Generated README consistency	✅ Passed	Passed · logs
Skill validation	✅ Passed	Passed · logs
Skill lint (vally)	✅ Passed	Passed · logs
Risk scan	✅ Passed	Passed · logs
Contributor reputation	✅ Passed	Passed · logs
Duplicate resource scan	✅ Passed	Passed · logs
PR quality signal	⏭️ Skipped	Skipped by its workflow · logs

Review

• Approvals: 0/1
• Assigned reviewer: not assigned yet — comment /request-review to ask for one
• Review target date: not set
• Still needed: 1 more approval(s)
• No staffed reviewer pool owns these files yet; any reviewer with write access counts as the domain reviewer.

Commands

Command	Who	What it does
/rerun-checks	PR author, maintainers	Re-runs failed or incomplete checks and re-evaluates this gate
/request-review	PR author, maintainers	Asks the review rotation to assign a reviewer (adds needs-reviewer)

_{Updated for 87e7f5b · gate run · This comment is maintained automatically — see submission gate docs.}

[github-actions]

🔍 Vally Lint Results

✅ All checks passed

Scope	Checked
Skills	1
Agents	0
Total	1

Severity	Count
❌ Errors	0
⚠️ Warnings	0
ℹ️ Advisories	0

Summary

Level	Finding
ℹ️	✅ recap (2/2 checks passed)
ℹ️	✓ [spec-compliance] All 1 skill(s) are spec-compliant.
ℹ️	✓ spec-compliance: All spec checks passed.
ℹ️	✓ [valid-refs] All file references across 1 skill(s) are valid.
ℹ️	✓ valid-refs: All file references resolve to existing files within the skill directory.
ℹ️	1 skill(s) linted, 1 passed

Full linter output

### Linting skills/recap
✅ recap (2/2 checks passed)
    ✓ [spec-compliance] All 1 skill(s) are spec-compliant.
        ✓ spec-compliance: All spec checks passed.
    ✓ [valid-refs] All file references across 1 skill(s) are valid.
        ✓ valid-refs: All file references resolve to existing files within the skill directory.

1 skill(s) linted, 1 passed

[Copilot]

🔵 Needs a closer look

Maintainer judgment is needed because the prose-only skill substantially overlaps with native session summarization, and its explicit /recap invocation and strict output contract may not provide sufficient repository-specific uplift.

No open issues 🎉

What changed in this PR

Adds an explicitly invoked /recap skill that provides concise, read-only summaries of the current Copilot session using only available conversation context.

Changes:

• Added the recap skill with strict output, context, and safety constraints.
• Registered the skill in the generated skills catalog.
• Added no scripts, dependencies, plugins, or hooks.

This pull request is ready to be approved.

File	Description
skills/​recap/​SKILL.md	Defines the on-demand, tool-free session recap behavior, metadata, and output limits.
docs/​README.skills.md	Registers the new recap skill in the generated catalog.

_{Review effort: Lite (auto)}

[mayankb2103]

/request-review

[github-actions]

🙋 Added needs-reviewer. Reviewer routing is assigning a reviewer now.

[mayankb2103]

Following up on the Copilot review: I agree that summarization itself is already a native model capability. The intended value of this skill is a repeatable, user-facing reorientation command, rather than a new summarization capability.

How this differs from native session summarization

• /compact is context management. It summarizes older conversation history and replaces that history in the model context to free space for continued work. Automatic compaction serves the same purpose.
• /recap is a quick reminder for the user. It returns only 1-2 short sentences, capped at 400 characters, covering the session goal and meaningful progress, plus established pending work or blockers when relevant. It does not request compaction, clear history, or resume unfinished work.
• An ordinary summary prompt can achieve a similar result. This skill packages the desired behavior into a reusable, explicitly invoked command with consistent brevity and scope: current conversation context only, no tool calls, file inspection, cross-session searches, or modifications. These are instructions to the assistant, not a separate runtime-enforced security boundary.

Use case

When returning to a terminal after an interruption, or switching between several ongoing Copilot sessions, a user can type /recap to answer "What were we doing here, and where did we leave off?" without scrolling through a long transcript or compacting the conversation. Once installed as a personal skill, the same command is available across projects.

Parallel with Claude Code

This deliberately parallels Claude Code's built-in /recap, which generates an on-demand one-line session summary capped at 400 characters. This contribution provides the on-demand reminder workflow as a Copilot skill; it does not attempt to reproduce Claude Code's automatic inactivity-triggered recap UI.

References: Copilot context management and compaction, Claude Code session recap.

I recognize that maintainers still need to decide whether this workflow convenience and explicit output contract provide enough differentiation for the collection.

[Copilot]

🟡 Changes recommended

A critical active finding remains because the recap source boundary could expose hidden system, developer, tool, or skill instructions instead of limiting summaries to user-visible conversation turns.

1 open issue

• Restrict recap source to user-visible conversation turns · New

_{Review effort: Lite (auto)}

[Copilot]

🔵 Needs a closer look

The skill uses an unsupported disable-model-invocation field, so its advertised explicit-only activation behavior is not guaranteed and the contract or implementation needs revision.

No open issues 🎉

1 previously missed issue

In code that hasn't changed since last review

Unsupported skill flag cannot enforce explicit-only activation

skills/​recap/​SKILL.md:4

disable-model-invocation is not a field in the Agent Skills specification, and Copilot CLI documents skills as being selected from the prompt and description; this flag is documented for custom agents rather than skills. It therefore cannot guarantee the advertised on-demand behavior: the skill may still be activated for unrelated session-summary requests. Please use a supported skill-level invocation control, or revise the contract/description to reflect that explicit-only activation cannot be enforced here.

1 resolved since last review

• Restrict recap source to user-visible conversation turns

_{Review effort: Lite (auto)}

[mayankb2103]

Regarding the latest review, the claim that disable-model-invocation is unsupported for Copilot skills appears to conflict with the current official CLI reference.

The Skill frontmatter fields section explicitly documents:

Field	Type	Required	Description
user-invocable	boolean	No	Whether users can invoke the skill with /SKILL-NAME. Default: true.
disable-model-invocation	boolean	No	Prevent the agent from automatically invoking this skill. Default: false.

These are documented under skills, not just custom agents. The same fields can be verified in the official documentation source.

For Copilot CLI, disable-model-invocation: true, together with the default user-invocable: true, therefore expresses the intended manual-only /recap behavior. The identical personal skill was also successfully invoked with /recap in Copilot CLI 1.0.105.

There is an important portability distinction: a field being absent from the core Agent Skills specification does not mean that a particular host does not support it. This contribution relies on documented Copilot CLI behavior; it does not claim that every Agent Skills-compatible host enforces this extension, nor that prompt instructions are a runtime security boundary.

Could this finding be reconsidered against the documented skill-level invocation controls?

[Copilot]

🟢 Approval recommended

The review record identifies no unresolved issues with the added /recap skill or its generated catalog entry.

No open issues 🎉

_{Review effort: Lite (auto)}

[Copilot]

🟡 Changes recommended

The recap skill is still a generic wrapper without a repository-specific workflow or other meaningful behavior change, so it needs a concrete differentiator.

1 open issue

• Skill adds no value beyond generic conversation summarization · New

_{Review effort: Lite (auto)}