CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

gsd-extract-learnings

Extract decisions, lessons, patterns, and surprises from completed phase artifacts

Quality

73%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide
SecuritybySnyk

Passed

No findings from the security scan

Fix and improve this skill with Tessl

tessl review fix ./pi/package-skills/gsd-extract-learnings/SKILL.md
SKILL.md
Quality
Evals
Security

Quality

Content

78%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.

This is a well-built adapter skill: it concisely conveys Pi-specific dispatch and include mechanics with explicit fail-closed error handling, and delegates all task detail to a single clearly signaled external workflow file one level deep. Its main limitations are inherited from its adapter nature — the real workflow is external and templated, and dispatch failures stop rather than recover — plus minor duplication of the workflow path.

DimensionReasoningScore

Conciseness

The body is lean: every paragraph carries non-obvious Pi-specific operational facts (no subagent API, gsd_dispatch semantics, fail-closed rules) with no explanation of concepts Claude already knows. Minor trimming is possible — the workflow path is stated both in <execution_context> and again in the closing line, and the two-sentence <arguments> explanation could be tightened — so it does not reach the every-token-earns-its-place bar of 5.

4 / 5

Actionability

The guidance is concrete and executable for an instruction-only skill: call gsd_dispatch with role and task, map Agent(...) calls via subagent_type/prompt, and follow explicit failure branches (STOP and surface; optional step → offer to skip; required step → stop the run). It is not 5 because the actual task workflow lives in an external, templated file (@__GSD_PI_ENGINE_ROOT__/gsd-core/workflows/extract-learnings.md) that is not part of the bundle, leaving a gap in what can be executed from the skill alone.

4 / 5

Workflow Clarity

A clear ordered sequence is present: read every file under <execution_context> "before doing anything else", treat <arguments> as data, then execute the workflow end-to-end, with explicit fail-closed checkpoints for dispatch failures. It is not 5 because there is no fix-and-retry feedback loop and no explicit handling for a missing or unreadable include file, only stop-and-surface semantics.

4 / 5

Progressive Disclosure

The body is a thin, well-organized adapter (~25 lines) whose only detailed content is a single one-level-deep reference, clearly signaled in a dedicated <execution_context> block with an explicit read-first instruction and repeated in the closing directive. No bundle files exist to inline, nothing belongs in a separate file within the skill, and navigation is trivial, matching the anchor for well-signaled one-level-deep references.

5 / 5

Total

17

/

20

Passed

Description

53%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.

The description communicates a clear and concrete "what" — extracting decisions, lessons, patterns, and surprises from completed phase artifacts — but it entirely lacks a "when to use" clause and natural trigger phrases, capping completeness and trigger term quality. Adding explicit usage triggers with common synonyms would substantially improve discoverability and correct-skill firing.

Suggestions

Add an explicit 'Use when...' clause naming trigger scenarios, e.g. "Use when a GSD phase completes or when the user asks for lessons learned, a retrospective, or a debrief."

Include natural synonyms users would actually say — "lessons learned", "retrospective", "debrief" — instead of relying only on GSD jargon like "phase artifacts" and "surprises".

Mention the concrete output (LEARNINGS.md) and the source artifacts (PLAN.md, SUMMARY.md, VERIFICATION.md, UAT.md, STATE.md) to sharpen the "what" and improve keyword coverage.

DimensionReasoningScore

Specificity

"Extract decisions, lessons, patterns, and surprises from completed phase artifacts" names one concrete action with a specific object list and source, matching the anchor for 1-2 concrete actions that is not comprehensive. It is not 4 because only a single action verb is present rather than several specific actions, and not 2 because it is far more concrete than a domain name with generic actions.

3 / 5

Completeness

The "what" is clear (extract decisions, lessons, patterns, and surprises from phase artifacts), but there is no "Use when..." clause or equivalent explicit trigger guidance; the "when" is only weakly implied by "completed", which caps this dimension at 3 per the judging guidelines. It is not 4 because the when-clause is entirely absent rather than merely imprecise.

3 / 5

Trigger Term Quality

Terms like "decisions", "lessons", and "patterns" are relevant keywords, but the description misses common natural variations users would say ("lessons learned", "retrospective", "debrief") and leans on GSD-specific jargon ("phase artifacts", "surprises"), matching the anchor for some relevant keywords missing synonyms. It is not 4 because several natural trigger phrases are absent, and not 2 because the keywords present are more than one or two generic terms.

3 / 5

Distinctiveness Conflict Risk

"completed phase artifacts" combined with the specific learning categories carves a mostly distinct niche with only minor overlap risk against closely related GSD extraction/documentation skills. It is not 5 because the absence of explicit trigger phrases leaves some ambiguity about when it should fire relative to sibling skills.

4 / 5

Total

13

/

20

Passed

Validation

100%

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

Validation — 16 / 16 Passed

Validation for skill structure

No warnings or errors.

Repository

open-gsd/gsd-core

Path

pi/package-skills/gsd-extract-learnings/SKILL.md

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.