CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

apm-3-initiate-worker

Initiate an APM Worker. Use only when the User manually invokes this skill.

Quality

76%

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

Fix and improve this skill with Tessl

tessl review fix ./templates/skills/apm-3-initiate-worker/SKILL.md
SKILL.md
Quality
Evals
Security

Quality

Content

78%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.

This is a well-structured instruction-only skill: it explains a novel multi-agent framework without padding, gives concrete file-level and step-level guidance, and keeps itself an overview by delegating detail to named guides one level deep. The main improvements are closing the few unspecified edge branches (missing bus files, agent ID delivery format) and eliminating the small amount of repeated phrasing. No destructive or batch operations are involved, so no workflow-clarity cap applies.

Suggestions

Add a short branch for §1.1 step 3 failure, e.g., "If a file is missing, tell the User which one and wait — do not create bus files yourself."

State how the agent ID is delivered (e.g., as an argument or in the invoking prompt) so registration is fully executable.

Deduplicate the trailing "wait for the User to invoke apm-4-check-tasks" instruction by stating it once at the end of §1.

DimensionReasoningScore

Conciseness

The body is lean and stepwise, and its framework explanation (Planner/Manager/Worker split, the bus) is novel information Claude does not already know rather than padding. Minor trimming is possible: "wait for the User to invoke apm-4-check-tasks" repeats in both branches of §1.1/§1.2, and "Otherwise, stop" restates the frontmatter restriction.

4 / 5

Actionability

Guidance is concrete and executable for an instruction-only skill: named files (".apm/bus/", "task.md", "report.md", "handoff.md", "AGENTS.md"), named sibling skills, and numbered steps with explicit conditionals ("If handoff.md has content, continue with §1.2"). Minor gaps keep it below 5: no branch for what to do when the bus folder is missing one of its three files, and how the agent ID is delivered is not specified.

4 / 5

Workflow Clarity

Sequencing is clear with genuine checkpoints: the three-file bus-folder check, checking each Task Prompt's Worker name against the registered identity, and reconciling Handoff claims against task.md and git with an explicit trust rule ("Where they differ, trust the current files and git"). Validation of the work itself is mandated but delegated to the task-execution guide, and there are no error-recovery branches, so it does not reach the level-5 anchor.

4 / 5

Progressive Disclosure

The body stays an overview and pushes detail to clearly labeled, one-level-deep guides ("{GUIDE_PATH:task-execution}: how to carry out a Task", "{SKILL_PATH:apm-communication}") with an explicit depth limit ("Read other APM files only when these files point you to them"). Sections are well organized and easy to navigate, and no bundle files exist that should have been split out.

5 / 5

Total

17

/

20

Passed

Description

62%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 is concise, precise, and answers both what the skill does and when to use it, with an explicit usage restriction appropriate for a manually invoked skill. It is held back by minimal action coverage and thin natural-language trigger terms. Adding one or two concrete actions (e.g., register on the bus, resume from handoff) and a few user-side trigger phrases would lift it.

Suggestions

List 1-2 concrete sub-actions after the lead sentence, e.g., "Initiate an APM Worker: register its bus folder, load Task Logs after a handoff, and enter the Task Prompt cycle."

Add natural trigger phrasing users would say, e.g., "Use when the user asks to start, set up, or resume an APM Worker session."

Keep the manual-invocation restriction but pair it with the specific invocation contexts (fresh initiation vs. continuing after a handoff) to sharpen distinctiveness from sibling APM skills.

DimensionReasoningScore

Specificity

"Initiate an APM Worker" names a specific domain and one concrete action, matching the anchor for 1-2 concrete actions that are not comprehensive. It lacks the several specific actions of a 4, while being more concrete than the purely generic single action of a 2.

3 / 5

Completeness

Both what ("Initiate an APM Worker") and when ("Use only when the User manually invokes this skill") are explicitly stated. The when clause is a single manual-invocation condition rather than concrete trigger phrases, so it falls short of the level-5 anchor.

4 / 5

Trigger Term Quality

Terms like "APM Worker", "initiate", and "manually invokes" are relevant to the skill, but common natural variations (start, set up, spawn a worker) are absent, matching the anchor for some relevant keywords missing common variations.

3 / 5

Distinctiveness Conflict Risk

"APM Worker" carves out a clear niche and the manual-invocation restriction limits mis-triggering, but the skill belongs to a family of closely related APM role skills, leaving minor overlap risk consistent with the level-4 anchor.

4 / 5

Total

14

/

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

sdi2200262/agentic-project-management

Path

templates/skills/apm-3-initiate-worker/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.