reflection-coach

Reflect on another agent's recent execution record and propose the smallest durable instruction, skill, or tool-description change. Use for evidence-backed coaching proposals, never hot-swaps.

Quality

74%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Low

Low-risk findings worth noting

Fix and improve this skill with Tessl

tessl review fix ./packages/skills-catalog/catalog/bundled/paperclip-operations/reflection-coach/SKILL.md

Quality

Content

81%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 highly actionable, well-sequenced procedure skill with strong validation/gating and a useful self-check checklist. Its main weakness is redundant restatement of the same invariants across several sections, which costs tokens without adding new guidance.

Suggestions

State each load-bearing invariant (no-same-run-apply, trajectories-not-scores, smallest-change) once in Hard guardrails and reference it by name elsewhere instead of restating it in the intro, Pitfalls, and Verification.

Consider moving the server-enforced mutation target keys and the curl endpoint catalog into a short reference section or file so the Procedure reads as pure sequence.

Trim the 'Two load-bearing rules' lead line since both rules are immediately expanded in Hard guardrails.

Dimension	Reasoning	Score
Conciseness	The body is dense and assumes Claude's competence (no basic-concept explanations), but the load-bearing invariants — 'no same-run apply', 'trajectories not scores', 'smallest change' — are restated across the intro, 'Two load-bearing rules', Hard guardrails, Procedure, Pitfalls, and Verification, which could be tightened.	3 / 5
Actionability	Fully executable guidance throughout: copy-paste curl commands with env vars and endpoints, an exact proposal-document template, enumerated server-enforced mutation target keys, and concrete skill-sync payload modes — specific examples cover the common cases.	5 / 5
Workflow Clarity	A clearly numbered 10-step sequence with explicit validation checkpoints (benchmark-gate in step 8, acceptance interaction gate in step 9, separate apply run in step 10), feedback loops (drop/reword rules that break replay; drop already-present rules), and a closing verification checklist.	5 / 5
Progressive Disclosure	Well-organized into clear sections (When to use / not to use, Inputs, Hard guardrails, Procedure, Pitfalls, Verification) with no nested references and no bundle files, but the long monolithic procedure with embedded API reference and mutation-target-key tables could arguably live in a separate reference file.	4 / 5
	Total	17 / 20 Passed

Description

67%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 specific, third-person description that names concrete actions and surfaces plus an explicit use clause and a distinguishing negative ('never hot-swaps'). It is held back from the top band by somewhat abstract trigger phrasing and missing common synonyms like 'review'.

Suggestions

Add a 'Use when...' clause naming concrete user-mentionable triggers (e.g. 'when asked to reflect on, review, or coach a specific agent's recent work') to lift completeness and trigger coverage.

Include natural synonyms users actually say — 'review', 'audit', 'agent performance' — alongside 'reflect' and 'coaching' to improve trigger term quality.

Optionally surface the evidence-backed / no-hot-swap distinction as an explicit trigger condition so it is less likely to fire for ad-hoc mid-session tweaks.

Dimension	Reasoning	Score
Specificity	Names the domain (coaching another agent) and concrete actions — 'Reflect on another agent's recent execution record' and 'propose the smallest durable instruction, skill, or tool-description change' — enumerating three change surfaces, but stops short of the comprehensive multi-action list of a 5.	4 / 5
Completeness	Clearly states what ('Reflect... and propose the smallest durable... change') and provides an explicit trigger clause ('Use for evidence-backed coaching proposals, never hot-swaps'), but the 'when' is somewhat abstract rather than concrete user-mentionable triggers, keeping it below a 5.	4 / 5
Trigger Term Quality	Includes relevant natural terms ('reflect', 'coaching', 'agent', 'execution record') but misses common synonyms a user might say such as 'review', 'audit', or 'performance', so coverage is partial rather than comprehensive.	3 / 5
Distinctiveness Conflict Risk	The framing — another agent's execution record, durable instruction/skill/tool-description change, 'never hot-swaps' — carves a clear niche with mostly distinct triggers, with only minor overlap risk against general code-review or agent-management skills.	4 / 5
	Total	15 / 20 Passed

Validation

87%

Warnings & errors only

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

Validation — 14 / 16 Passed

Validation for skill structure

Criteria	Description	Result
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning
relative_links	Relative link issues: 4 suspicious	Warning

	Total	14 / 16 Passed

Repository: paperclipai/paperclip
Path: packages/skills-catalog/catalog/bundled/paperclip-operations/reflection-coach/SKILL.md
Commit: 890ab9a

Reviewed: about 17 hours ago

Table of Contents

Discovery Implementation Validation

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.