CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

handoff

Safely commit, verify, and push completed work on a feature branch in this repo, then open a PR and record the review as a PR comment. Use when work is finished and you want to hand it off — it runs the repo's verification gate, preserves unrelated work, and never touches protected branches without the required confirmation.

Quality

86%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide
SecuritybySnyk

Low

Low-risk findings worth noting

SKILL.md
Quality
Evals
Security

Quality

Content

88%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.

Excellent operational skill content: exact commands, named scripts, explicit validation checkpoints, and error-recovery guidance make it directly executable. The only real weaknesses are mild verbosity in the step-1 caution detail and dated incident references inlined in the main flow rather than separated out.

Suggestions

Move the 'Restarting a branch whose PR already merged' anecdote (with its 2026-08-14 date and PR #1944 reference) into a short 'Known pitfalls' or reference section so dated incident details don't sit in the main workflow.

Tighten step 1 by consolidating the three repeated snapshot/compare instructions into one 'record → re-check → compare → stop on any drift' pattern stated once.

Consider splitting the detailed preconditions for shared worktrees into a referenced file (e.g. references/shared-worktree-safety.md) to keep the main sequence lean.

DimensionReasoningScore

Conciseness

The body is dense with repo-specific operational detail and assumes git competence — no space is spent explaining concepts Claude already knows. Minor trimming is possible (step 1 repeats the snapshot-compare instructions), and dated incident references ('Observed 2026-08-14', 'PR #1782', 'PR #1944') are time-sensitive details not placed in an old-patterns section, keeping it below the lean-everywhere anchor.

4 / 5

Actionability

Fully executable guidance throughout: exact commands ('git status --porcelain=v1 -z --untracked-files=all', 'npm run verify:pr-local', 'gh pr create --base main', 'npm run pr:areas'), named scripts with paths ('node scripts/check-base-freshness.mjs', 'scripts/guard-push.mjs'), and all five guard override env variables listed. Copy-paste ready and covering the common cases.

5 / 5

Workflow Clarity

A clear 9-step sequence with explicit validation checkpoints (record/compare the cached index before staging, verify the commit with 'git show --name-only --format=fuller HEAD', confirm the remote tip with 'git ls-remote', 'Paste the decisive proof line') and error-recovery loops (stale ref after merge → 'git remote prune origin' then push; heed guard blocks rather than overriding). This matches the top anchor including feedback loops.

5 / 5

Progressive Disclosure

No bundle files exist, so all content is in SKILL.md; sections are well organized (Preconditions, numbered Steps, Requires explicit confirmation) and repo-doc links are clearly signaled with anchors. However, the ~100-line body inlines material that could sit in a reference file — the merged-branch restart anecdote and the long step-1 caution block — so structure is good but not cleanly split, and the under-50-line simple-skill exception does not apply.

4 / 5

Total

18

/

20

Passed

Description

80%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: concrete multi-action capability statement plus an explicit 'Use when...' clause and a distinctive safety-scoped niche. Its only weaknesses are the general (rather than trigger-phrase-specific) when-clause and a few missing natural keyword variants like 'pull request' spelled out.

DimensionReasoningScore

Specificity

Lists multiple concrete actions covering the skill's full scope: 'commit, verify, and push completed work on a feature branch', 'open a PR', 'record the review as a PR comment'. This matches the comprehensive multi-action anchor; the score below (4) would suit a list with coverage gaps, which this does not have.

5 / 5

Completeness

Explicitly answers both: the what ('Safely commit, verify, and push... then open a PR and record the review as a PR comment') and the when ('Use when work is finished and you want to hand it off'). The when-clause is explicit but general rather than concrete trigger phrases, so it fits anchor 4, not 5.

4 / 5

Trigger Term Quality

Natural phrases users would say are present — 'commit', 'push', 'open a PR', 'feature branch', 'hand it off', 'work is finished' — but common variants like 'pull request' spelled out or 'create a PR' are missing. Good coverage, just short of the comprehensive-synonyms anchor.

4 / 5

Distinctiveness Conflict Risk

'This repo's safe Git handoff... verification gate... never touches protected branches' carves a distinct niche unlikely to fire for unrelated skills. Minor overlap remains with generic git or commit-message skills, so it is mostly distinct rather than minimal-conflict.

4 / 5

Total

17

/

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

relative_links

Relative link issues: 4 suspicious

Warning

referenced_paths_exist

Referenced path issues: 2 possibly missing (scripts/check-base-freshness.mjs, scripts/guard-push.mjs)

Warning

Total

14

/

16

Passed

Repository

BigSimmo/PsychSift

Path

.claude/skills/handoff/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.