CtrlK
BlogDocsLog inGet started
Tessl Logo

address-feedback

Respond to an automated reviewer's findings on the current feature PR — a narrow, headless responder. Triages each reviewer finding into Clear / Ambiguous / Complex / Out-of-PRD-Scope, then acts — fixes the cause and pushes (diff-only) for Clear, posts one in-thread reply for the rest. Skips findings it already handled. Invoked headless by the dispatcher's feedback gate; the dispatcher owns the round counter and the STUCK escalation. Triggers - address-feedback, address review comments, respond to PR review, handle reviewer findings, PR feedback loop (project)

68

Quality

84%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

SecuritybySnyk

Low

Low-risk findings worth noting

The canonical home for this skill is address-feedback in tdg-ninja/context-specs-factory-ai

SKILL.md
Quality
Evals
Security

Quality

Content

85%Weight 40%Scale 1-5

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

A well-structured, actionable instruction skill with a clear sequenced workflow, explicit idempotency validation, a hard-nevers checklist, and clean one-level-deep progressive disclosure. The only gaps are minor verbosity in the rationale section and command detail living in references rather than inline.

Suggestions

Tighten the 'one-way model (read this first)' section to its load-bearing consequences, trimming the philosophical framing.

Consider inlining the one or two most-used gh commands (e.g., listing unresolved reviewer findings) so the core pass is executable without opening the reference.

Dedupe the never-silence / skip / no-refactor rules between the body sections and 'Hard nevers' so the checklist stays the single source.

DimensionReasoningScore

Conciseness

Lean and assumes Claude's competence — no generic explanation of PRs or code review — with deliberate checklist reinforcement in 'Hard nevers'; not a 5 because the 'one-way model' rationale section and repeated rule statements could be trimmed further.

4 / 5

Actionability

Gives concrete per-bucket actions and enumerates specific forbidden patterns (eslint-disable*, @ts-ignore, .skip, xfail, etc.); not a 5 because the exact executable gh/GraphQL commands are deferred to the reference file rather than present inline.

4 / 5

Workflow Clarity

Clear numbered sequence (re-derive → triage → act → exit) with explicit validation checkpoints (reply-idempotency, fix-idempotency), a feedback loop (re-derive failures from the worktree → no diff → no commit), and a 'Hard nevers' checklist for the destructive push/reply operations.

5 / 5

Progressive Disclosure

Overview body offloads fiddly detail to two clearly signaled, one-level-deep, verified files — `references/gh-mechanics.md` for commands and `references/triage.md` for the taxonomy — with no nested references and easy navigation.

5 / 5

Total

18

/

20

Passed

Description

83%Weight 40%Scale 1-5

Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.

A tight, specific, third-person description that names concrete actions, a clear niche, and explicit trigger phrases. It is held back only by a 'when' clause that frames invocation through the dispatcher mechanism rather than a user-facing 'Use when...' phrasing.

Suggestions

Reframe the 'when' as a user-facing 'Use when responding to an automated reviewer's findings on a feature PR' clause so completeness can reach 5.

Add a couple of natural trigger variants (e.g., 'fix review findings,' 'reply to reviewer comments') to broaden trigger-term coverage.

Lead with the trigger phrases before the dispatcher-mechanism detail so the user-facing intent comes first.

DimensionReasoningScore

Specificity

Lists multiple concrete actions — triages into four named buckets, 'fixes the cause and pushes (diff-only) for Clear, posts one in-thread reply for the rest,' and 'Skips findings it already handled' — giving comprehensive coverage of what it does.

5 / 5

Completeness

Both 'what' and 'when' are present and concrete, with an explicit 'Triggers -' line; not a 5 because the 'when' leans on the dispatcher invocation mechanism rather than a cleanly integrated user-facing 'Use when...' clause.

4 / 5

Trigger Term Quality

Trigger line supplies natural phrases users would say ('address review comments,' 'respond to PR review,' 'handle reviewer findings,' 'PR feedback loop') with good synonym coverage, but all rotate around one concept and a few natural variants are absent.

4 / 5

Distinctiveness Conflict Risk

'Respond to an automated reviewer's findings on the current feature PR — a narrow, headless responder' carves a clear niche with distinct triggers and minimal overlap with other skills.

5 / 5

Total

18

/

20

Passed

Validation

93%

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation15 / 16 Passed

Validation for skill structure

CriteriaDescriptionResult

referenced_paths_exist

Referenced path issues: 2 missing

Warning

Total

15

/

16

Passed

Repository
tdg-ninja/context-specs-claude-code
Reviewed

Table of Contents

Is this your skill?

If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.