CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

reflect-session

Reflect on the current session, extract learnings, and consolidate them into Hindsight for future recall.

Quality

81%

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

SKILL.md
Quality
Evals
Security

Quality

Content

100%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 lean, fully executable instruction skill: a clear six-step sequence with user-approval and confirmation checkpoints, a concrete learning taxonomy, and a copy-paste output template. It assumes Claude's competence, wastes no tokens on known concepts, and appropriately needs no external reference files.

DimensionReasoningScore

Conciseness

The body is lean and assumes Claude's competence: a one-line framing ('You are wrapping up a session'), five reflection questions, a six-category taxonomy, a copy-paste output template, and three terse guardrails ('Be honest about what went wrong', "Don't retain trivial things", "don't force it"). Nothing explains concepts Claude already knows, and every section earns its place, matching anchor 5.

5 / 5

Actionability

Fully executable instruction-only guidance: numbered steps, a concrete learning taxonomy (Mental Model, Hard Rule, Gotcha, Decision, Technique, Discovery), an exact output template, the named Hindsight 'retain' tool, an explicit default ('Default is all of them unless the user removes some'), and a confirmation step. Per the scoring notes, absence of code is not penalized since the guidance is directly actionable — this fits anchor 5's 'copy-paste ready' standard.

5 / 5

Workflow Clarity

Six steps in a clear sequence with explicit checkpoints: user approval of learnings before retention (step 4, 'Ask the user which learnings to retain') and verification of what was retained (step 6, 'Confirm the number of memories retained and list their titles'). Retention is user-gated, so the batch-operation validation cap does not apply, and this matches anchor 5's explicit validation and feedback structure.

5 / 5

Progressive Disclosure

This is a simple, single-purpose, self-contained skill (~45 non-blank body lines) with no external detail that needs splitting out, organized into clear sections (## Steps, ## Important) with no nested references. Per the simple-skill exception, well-organized sections with no need for external references merit anchor 5.

5 / 5

Total

20

/

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 the skill's purpose clearly and names its target system (Hindsight), but it lacks any explicit 'Use when...' trigger guidance, which caps completeness and limits trigger-term coverage. Its verbs are procedural ('reflect', 'consolidate') rather than concrete capabilities, leaving it distinguishable but under-specified relative to the strong reference examples.

Suggestions

Add an explicit trigger clause, e.g. 'Use when wrapping up a session, when the user asks for a retrospective or to reflect on what was learned, or at the end of a long task.'

Include natural synonyms and phrases users would actually say — 'retrospective', 'wrap-up', 'debrief', 'save learnings' — to improve trigger-term coverage.

Enumerate concrete capabilities (categorize learnings, present them for user review, store them via the Hindsight retain tool) instead of abstract phases like 'reflect' and 'consolidate'.

DimensionReasoningScore

Specificity

The description names three actions ('Reflect on the current session, extract learnings, and consolidate them into Hindsight for future recall') and a concrete target system (Hindsight), but the actions are procedural phases rather than the distinct concrete capabilities of the anchor-4 examples, and it omits real capabilities such as categorizing learnings and presenting them for user review. It fits anchor 3 — a clear domain with a couple of actions, not comprehensive — better than anchor 4.

3 / 5

Completeness

The 'what' is clear (reflect on the session, extract learnings, consolidate into Hindsight), but there is no 'Use when...' clause or equivalent explicit trigger guidance, which caps completeness at 3 per the judging guidelines. 'When' is only weakly implied by the phrase 'the current session', fitting anchor 3 exactly; it cannot be 4 without an explicit 'when'.

3 / 5

Trigger Term Quality

It includes some natural terms users would say ('reflect', 'session', 'learnings', 'recall') but misses common variations and synonyms such as 'retrospective', 'wrap-up', 'debrief', or 'save what we learned'. This matches anchor 3 (some relevant keywords, missing common variations) rather than anchor 4's good-but-gapped coverage.

3 / 5

Distinctiveness Conflict Risk

The named 'Hindsight' system and session-reflection framing give it a distinct niche with only minor overlap risk against generic memory/retention skills, matching anchor 4. It is not a 5 because it lacks explicit distinct trigger phrases that would fully separate it from other memory-saving 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

AndreJorgeLopes/devflow

Path

skills/reflect-session/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.