Tessl • The management layer for AI agent skills

SkillPublic

thread-state

Move a tracked AI work thread to a different lifecycle stage on the thread registry/board. Use when the auto-classifier put a thread in the wrong stage (Funnel, Specify/Plan, Implement, Review/Ship, Waiting/Blocked, Done) and you want to correct it, or to manually advance a thread.

Quality

88%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Passed

No findings from the security scan

Quality

Content

93%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 an exemplary compact operational skill: fully executable commands, a clear two-path workflow with a uniqueness guardrail, and dense non-obvious domain facts with zero padding. The only minor gap is the absence of an explicit verification step after applying a stage change.

Dimension	Reasoning	Score
Conciseness	The body is lean and every token earns its place: it conveys only non-obvious operational facts ("stored as `manual_stage` and survives every rescan", "On Hold is a manual-only parking lane — the classifier never assigns it", per-harness command install paths, the Windows node fallback). No concepts Claude already knows are re-explained and there is no padding.	5 / 5
Actionability	Guidance is fully executable and copy-paste ready: "/thread-state done" with note-attachment syntax, "node scripts/edit-thread.mjs --list", "--match \"<id-or-title>\" --stage \"<stage>\"", "--here", and "--note", with loose stage-matching semantics explained. Both common cases (staging the current thread vs. another thread) are covered with concrete commands.	5 / 5
Workflow Clarity	Two clearly sequenced paths (quick path for the current thread; numbered workflow for another thread) with concrete commands and a built-in guardrail ("--match ... must match exactly one thread"). It stops short of an explicit post-change verification step (e.g., confirming the rendered board), so it matches the "clear sequence with most checkpoints" anchor rather than the full validation-loop anchor.	4 / 5
Progressive Disclosure	A compact (~65-line) single-file skill with well-organized sections (Outcome, When to use, Valid stages, Quick path, Workflow, Notes); nothing inlined that belongs in a separate file, and the few cross-references (skills/finish-thread, skills/track-thread, scripts/edit-thread.mjs) are clearly signaled and one level deep. No bundle files exist to split further.	5 / 5
	Total	19 / 20 Passed

Description

78%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 well-constructed description with an explicit what-and-when structure and domain-specific triggers. Its main weaknesses are a single-action capability statement weakened by second-person phrasing, and a stage list that omits "On Hold" and includes "Waiting/Blocked", which the skill body says is not a stage.

Suggestions

Align the enumerated stages with the board's real stages — replace "Waiting/Blocked" (a flag, not a stage, per the body) with "On Hold" — so trigger terms match what users would actually say.

Use third-person voice in the trigger clause (e.g., "Use when the auto-classifier put a thread in the wrong stage and the user wants to correct it") instead of "you want to correct it".

Add natural trigger synonyms such as "move a thread to done", "put a thread on hold", or "advance a thread to Review/Ship" to broaden keyword coverage.

Dimension	Reasoning	Score
Specificity	"Move a tracked AI work thread to a different lifecycle stage on the thread registry/board" names the domain and one concrete action with its target artifact, but does not list several distinct actions (correct/advance are purposes for the same operation). The second-person "you want to correct it" in the trigger clause further weakens the specificity per the third-person voice guideline.	3 / 5
Completeness	It explicitly answers both questions: the "what" (move a thread's lifecycle stage on the registry/board) and a clear "Use when..." clause with two concrete triggers (auto-classifier misclassified the stage; deliberate manual advancement). This matches the anchor for a complete what-and-when description.	5 / 5
Trigger Term Quality	Natural phrases like "put a thread in the wrong stage" and "manually advance a thread" plus stage names (Funnel, Specify/Plan, Implement, Review/Ship, Done) give good coverage, but common variations users would say ("move a thread to done", "on hold", "archive", "triage") are missing, and "Waiting/Blocked" as listed does not match the board's actual stage names.	4 / 5
Distinctiveness Conflict Risk	The niche (correcting lifecycle stages in an AI thread registry) and its triggers are clearly distinguishable from generic skills. Minor overlap risk remains with closely related sibling skills such as finish-thread (marking a thread done) and track-thread, which handle adjacent thread-registry operations.	4 / 5
	Total	16 / 20 Passed

Validation

93%

Warnings & errors only

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation — 15 / 16 Passed

Validation for skill structure

Criteria	Description	Result
referenced_paths_exist	Referenced path issues: 2 missing	Warning

	Total	15 / 16 Passed

Repository: yyeret/ai-threads-kanban
Path: skills/thread-state/SKILL.md
Revision: 9f94822

Reviewed: 6 days ago

Discovery Implementation Validation

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.

thread-state

Table of Contents

Is this your skill?