CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

retain-learning

Retain a new learning, discovery, or hard-won insight into Hindsight so it's available in future sessions.

Quality

77%

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 ./devflow-plugin/skills/retain-learning/SKILL.md

The canonical home for this skill is retain-learning in AndreJorgeLopes/devflow

SKILL.md
Quality
Evals
Security

Quality

Content

92%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.

The body is a tight, well-structured instruction skill: a clear five-step workflow with explicit input handling, a concrete category taxonomy, defined output fields, and a complete worked example. Its only weakness is that the final storage step does not specify the exact Hindsight retain tool invocation format.

Suggestions

Specify the exact Hindsight `retain` tool call — parameter names and how each structured field (Title, Category, Content, Context, 'Why it matters') maps to the payload — so memories are stored consistently.

Add a second worked example covering a different category (e.g., a Gotcha or Mental Model) to demonstrate how classification varies.

DimensionReasoningScore

Conciseness

43 lean lines: a 5-step workflow, a compact category taxonomy, field specs, and one tight worked example. No padding or concept re-teaching; every line instructs ('Title: A concise, searchable title (max 10 words)'). Matches the score-5 anchor — lean, efficient, every token earns its place.

5 / 5

Actionability

Concrete classification taxonomy, explicit field structure, and a fully worked example make the guidance mostly executable. However, 'Store the memory using the Hindsight `retain` tool. Pass the structured content as the memory payload' never specifies the tool's actual parameters, a minor executable gap that keeps it at the score-4 anchor rather than 5.

4 / 5

Workflow Clarity

Clear 5-step sequence (parse, classify, structure, store, confirm) with explicit empty-input handling ('If no arguments are provided, ask the user what they learned') and a confirmation step showing what was retained. The operation is non-destructive, and as a simple single-purpose skill with an unambiguous flow it meets the score-5 anchor under the rubric's simple-skill exception.

5 / 5

Progressive Disclosure

Under 50 lines, single purpose, no external references needed; content is well-organized into '## Steps' and '## Example' sections with no inlined material that belongs elsewhere. Per the rubric's simple-skill guideline this is a score 5.

5 / 5

Total

19

/

20

Passed

Description

50%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 clearly states what the skill does and names its target system (Hindsight), but it is a single-action description with no explicit trigger guidance. Its main gaps are the missing 'Use when...' clause and the absence of natural synonyms users would actually say when invoking it.

Suggestions

Add an explicit 'Use when...' clause, e.g., 'Use when the user says "remember this", "save this learning", or asks to store a finding in Hindsight.'

Include natural trigger synonyms users would say — 'remember', 'save', 'note', 'takeaway' — rather than only 'learning, discovery, insight'.

Briefly state what Hindsight is or how retention differs from other memory/note-taking flows to reduce overlap risk with generic memory skills.

DimensionReasoningScore

Specificity

Names one concrete action ('Retain a new learning, discovery, or hard-won insight into Hindsight') plus the target system, but coverage is limited to that single action rather than several specific capabilities. This matches the score-3 anchor (domain and 1-2 concrete actions, not comprehensive); it is above 2 because it is not generic, and below 4 because no list of several specific actions is present.

3 / 5

Completeness

The 'what' is clear (retain learnings into Hindsight for future sessions) but there is no 'Use when...' clause or equivalent explicit trigger guidance — 'when' is only weakly implied. Per the rubric guideline, a missing trigger clause caps completeness at 3.

3 / 5

Trigger Term Quality

Includes relevant natural terms ('learning', 'discovery', 'hard-won insight') but misses the common variations and synonyms users would naturally say ('remember this', 'save this', 'note it', 'takeaway'). Matches the score-3 anchor (some relevant keywords, missing common variations); not 4 because natural-term coverage has clear gaps.

3 / 5

Distinctiveness Conflict Risk

Naming the specific system ('Hindsight') gives it some distinctness, but 'learning, discovery, or hard-won insight' territory overlaps broadly with generic memory/journaling skills and no distinct trigger phrases separate it. Between anchors 3 and 4; the overlap risk with closely related memory skills is more than minor, so it lands at 3.

3 / 5

Total

12

/

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

AndreJorgeLopes/devflow

Path

devflow-plugin/skills/retain-learning/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.