CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

finish-session

Use when the user asks to finish or wrap up the session — checks that all they asked for is delivered, then commits, pushes and archives it. Not for a code review. Self-contained — execute the inlined steps directly, do NOT invoke other skills.

Quality

79%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide
SecuritybySnyk

Critical

Do not install without reviewing

Fix and improve this skill with Tessl

tessl review fix ./plugins/flowai/skills/finish-session/SKILL.md
SKILL.md
Quality
Evals
Security

Quality

Content

63%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.

An exceptionally actionable, well-gated composite workflow undermined by serious redundancy and monolithic structure: one hand-off step is duplicated three times over, dated incident notes are embedded in runtime instructions, and everything lives in a single 45KB file with no progressive disclosure. The procedural core itself is copy-paste executable with thorough validation.

Suggestions

Fix the Push Phase's duplicated hand-off: step 7 appears twice (doubled heading "7. **7. **Hand off to the next phase") and its full text is also embedded parenthetically inside step 6's 'Absent' branch — collapse to one canonical step 7.

Deduplicate the doc-sync rename guidance: the "grep the OLD name, not the NEW name" warning appears nearly verbatim both in the mapping list and again under "Sync each affected document" — state it once.

Move dated incident annotations ("Measured 2026-08-17…", "2026-09-02 and 2026-09-06…") out of the runtime instructions into a maintainer note or separate change-log reference file.

DimensionReasoningScore

Conciseness

Noticeable copy-paste padding throughout: the Push Phase hand-off step appears twice (doubled heading "7. **7. **Hand off to the next phase") AND its full text is inlined parenthetically inside step 6's Absent branch; the "grep the OLD name" rename warning is stated nearly verbatim twice (Commit Phase step 2); dated incident reports ("Measured 2026-08-17", "2026-09-02 and 2026-09-06") are embedded inline. Not 3 because these are genuine duplicated sections, not merely tightenable prose; not 1 because the bulk is dense, non-obvious operational instruction with no explanation of concepts Claude already knows.

2 / 5

Actionability

Fully executable throughout: exact commands ("git status -s && echo '---DIFF---' && git diff..."), exact output formats ("Goals: all met (<N> of <N>)"), exact user prompts, and an exact task-status derivation rule ("K=0→\"to do\", 0<K<N→\"in progress\", K=N→\"done\""). The only abstract points (the host's archive tool) are necessarily host-dependent and explicitly scoped ("Decide from the tool list alone").

5 / 5

Workflow Clarity

Five strictly ordered phases with four explicit gates, STOP conditions everywhere, post-push verification ("git rev-parse @{u}" vs HEAD), pre-reset safety checks before "git reset --hard", and a final verification checklist — exemplary sequencing with feedback loops. Not 5 because the corrupted step 6/7 structure in the Push Phase (step 7's full text nested inside step 6's parenthetical, then repeated as a doubled heading) makes the hand-off genuinely ambiguous at that point; not 3 because validation checkpoints are pervasive, including for the destructive reset.

4 / 5

Progressive Disclosure

Well-organized internal structure (phases, gates, rules, verification checklist) but a monolithic ~380-line body with zero bundle files: the CI-await tunables machinery, the full doc-sync mapping, and the entire Reflect audit rubric are all inlined. The inlining is a deliberate, explicitly stated design choice ("Self-contained — execute the inlined steps directly, do NOT invoke other skills" — to preserve gate integrity), which keeps this at 3 rather than 2 despite content that would otherwise belong in separate reference files.

3 / 5

Total

14

/

20

Passed

Description

87%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 strong description: explicit "Use when" trigger in natural phrasing, a concrete what-clause, and a disambiguating exclusion. The only gaps are a missing synonym or two in the trigger terms and the unmentioned Reflect phase.

DimensionReasoningScore

Specificity

"checks that all they asked for is delivered, then commits, pushes and archives it" names four concrete actions (verify goals, commit, push, archive). Not 5 because the Reflect Phase — audit plus corrective edits to instruction files, roughly a fifth of the skill's body — is entirely unmentioned, leaving a real coverage gap.

4 / 5

Completeness

Explicitly answers both: what ("checks that all they asked for is delivered, then commits, pushes and archives it") and when ("Use when the user asks to finish or wrap up the session"), with concrete trigger phrases plus a disambiguating exclusion ("Not for a code review"). Matches the anchor-5 example's structure.

5 / 5

Trigger Term Quality

"finish or wrap up the session" is exactly what a user would naturally say. Not 5 because common variations ("end the session", "close out", "we're done for today") are missing; not 3 because the core natural trigger phrases are present and well chosen.

4 / 5

Distinctiveness Conflict Risk

A clear niche (session close-out) with a specific trigger phrase unlikely to fire for bare "commit" or "push" requests, and an explicit negative boundary ("Not for a code review") that further reduces mis-triggering. Minimal conflict risk; fits the anchor-5 example.

5 / 5

Total

18

/

20

Passed

Validation

87%

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

CriteriaDescriptionResult

frontmatter_unknown_keys

Unknown frontmatter key(s) found; consider removing or moving to metadata

Warning

referenced_paths_exist

Referenced path issues: 1 possibly missing (scripts/generate-skill-composites.ts)

Warning

Total

14

/

16

Passed

Repository

korchasa/flowai-plugins

Path

plugins/flowai/skills/finish-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.