CtrlK
BlogDocsLog inGet started
Tessl Logo

write-issue

Use when writing a GitHub issue in this repository — planned work, a bug, an epic and its children, a follow-up a review or retro turned up, or a needs:decision proposal. Covers which repository it goes to, the title, type and labels, a body someone can act on months later with nothing checked out, the faults agent-written issues fall into, and the command that files it. Invoked by new-feat, prep-pr, retro and stress-plan whenever they open an issue.

75

Quality

94%

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

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.

A tightly written, highly actionable instruction skill: concrete gh commands throughout, an explicit six-step workflow with a pre-flight checklist and a no-network fallback, and clean one-level-deep delegation to wiki conventions and an examples file. The only weakness is mild redundancy between Step 5's rules and the faults table.

DimensionReasoningScore

Conciseness

The body is dense and repo-specific with no generic filler — it never explains what an issue or gh is, and constraints like 'the local shell is fish, which has no heredocs' are concrete and load-bearing. It falls just short of anchor 5 because 'The faults, by name' largely restates Step 5's rules (stock paragraph↔rule 1, empty section↔rule 2, want-not-problem↔rule 4, slash shorthand↔rule 7, done-when-cannot-fail↔rule 5, pointer-only↔rule 8), a minor redundancy that could be merged into the rules as inline examples.

4 / 5

Actionability

Guidance is fully executable: copy-paste-ready commands with appropriate placeholders for the search ('gh issue list --repo <repository from the table> --state all --limit 20 --search ...'), label listing, both issue-create variants (public and tracker, with exact flag sets), and the full project-item workflow including the --jq filters. Concrete formats are given for titles and body lead-ins, and common cases (planned work, bug, epic, finding, needs:decision) each get specific instructions, with a worked example offloaded to references/examples.md.

5 / 5

Workflow Clarity

Steps 1–6 are clearly sequenced (routing → duplicate search → title → type/labels → body → filing) with explicit validation: the Step 1 duplicate search across open and closed issues, the pre-flight checklist before running gh ('Right repository... no open duplicate', 'Every path and line in the body exists today', '**Done when** could fail'), post-filing steps, and an error-recovery fallback for 'No network, no gh' that forbids claiming an issue was filed. This matches the anchor-5 pattern of explicit validation, feedback loops, and checklists.

5 / 5

Progressive Disclosure

The body is a clear overview that owns issue writing and explicitly delegates what it does not own ('Two things this skill does not own, and links to instead') to one-level-deep wiki paths, and to references/examples.md for worked examples — verified to exist and referenced once, clearly signaled ('references/examples.md has one'). Content is appropriately split and easy to navigate, matching anchor 5; no nested multi-level references exist.

5 / 5

Total

19

/

20

Passed

Description

92%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 dense, explicit description that clearly states both what the skill covers and when to use it, with concrete trigger phrases and named sibling-skill boundaries. The only minor gap is a few missing natural synonyms such as 'create an issue' or 'task'.

DimensionReasoningScore

Specificity

The description enumerates the skill's full concrete scope — 'which repository it goes to, the title, type and labels, a body someone can act on months later... the faults agent-written issues fall into, and the command that files it' — with concrete issue kinds (planned work, bug, epic and children, follow-up, needs:decision proposal). This matches the anchor 'lists multiple specific concrete actions; comprehensive coverage' and mirrors every major step present in the body, so it is not the level 4 anchor, which requires minor gaps in coverage.

5 / 5

Completeness

It explicitly answers both questions: 'Use when writing a GitHub issue in this repository — planned work, a bug, an epic and its children, a follow-up a review or retro turned up, or a needs:decision proposal' gives concrete when-triggers, and the 'Covers...' sentence states exactly what the skill does. This matches the anchor-5 example pattern (explicit what + explicit when with concrete trigger phrases), not the level-4 anchor where 'when' could be more specific.

5 / 5

Trigger Term Quality

Natural trigger phrases are strong — 'writing a GitHub issue', 'planned work', 'a bug', 'an epic and its children', 'a follow-up a review or retro turned up', 'needs:decision proposal', plus verb variants 'the command that files it' and 'whenever they open an issue'. A few natural terms a user might say are still absent ('create an issue', 'task', 'backlog item'), matching anchor 4 ('good keyword coverage; a few natural terms missing') rather than the fully comprehensive synonym coverage of anchor 5.

4 / 5

Distinctiveness Conflict Risk

It occupies a clear niche — writing GitHub issues in this repository — and even demarcates its boundaries against sibling skills ('Invoked by new-feat, prep-pr, retro and stress-plan whenever they open an issue'), which keeps it from firing for pull-request or review work. Minimal conflict risk, matching anchor 5; anchor 4's 'minor overlap risk with closely related skills' does not apply because the sibling relationships are explicitly named.

5 / 5

Total

19

/

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
englishstreetventures/osn
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.