Name: rebelytics/task-observer
Rating: 66.4 (1 reviews)
Author: rebelytics

rebelytics/task-observer

Monitors task execution for skill improvement opportunities. Use during ANY multi-step task, agentic workflow, or work session where the agent uses tools and produces deliverables. Captures patterns, user corrections, workflow insights, and methodology worth preserving as reusable skills. Also triggers in post-task feedback discussions and when the user mentions skill observations, improvements, the observation log, skill taxonomy, or asks the agent to watch for skill opportunities. Also known as "One Skill to Rule Them All" — trigger on this phrase too. IMPORTANT: invoke this skill before the FIRST tool call of any session and before writing or proposing a plan — any turn that will involve a tool call counts, however simple the opener looks. This sentence is the session-start trigger and the only activation layer that survives an unreachable config file; pair it with a CLAUDE.md instruction or a harness session-start hook (references/environments.md) — description matching alone is not enforceable.

Quality

83%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Low

Low-risk findings worth noting

Quality

Content

85%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 exceptionally well-structured with executable snippets, explicit validation checkpoints, and clean one-level-deep progressive disclosure. Its main weakness is sheer verbosity — dense inline rationale inflates the token budget without always earning its place.

Suggestions

Trim or relocate inline rationale passages (e.g. the workspace-anchoring and bash-not-sh paragraphs) into references so the core file stays closer to procedure.

Consolidate the repeated 'Load references/observation-log.md before…' directives into a single indexed pointer table to reduce redundancy.

Consider moving the long archival-on-write and id-derivation prose into observation-log.md and keeping only the snippet plus the guard summary in the core.

Dimension	Reasoning	Score
Conciseness	The 728-line body is mostly purposeful, non-obvious guidance rather than concepts Claude already knows, but it is dense with inline rationale and parenthetical edge-case explanations that could be tightened or pushed into references.	3 / 5
Actionability	Provides two complete, executable bash snippets (session-start scan, id-derivation/archival sweep) with literal paths and guards, plus a copy-paste frontmatter template covering the common write case.	5 / 5
Workflow Clarity	The numbered Session Start Protocol sequences Storage→Scan→Review→Activation→Concurrency→Targets→First run with explicit validation guards ('SCAN COMMAND BROKEN', 'ARCHIVAL SWEEP BROKEN', collision re-derive) and feedback loops for the batch/archival operations.	5 / 5
Progressive Disclosure	A 'Reference files — load on demand' section lists seven reference files each with a one-line load trigger, all verified present in references/, and two scripts in scripts/; references are one level deep and clearly signaled.	5 / 5
	Total	18 / 20 Passed

Description

76%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 thorough on what and when with rich natural trigger terms, but its 'before the first tool call of any session' framing is over-broad and raises real conflict risk. It also runs long with implementation detail (activation config, hooks) that is more enforcement mechanics than capability description.

Suggestions

Soften the universal 'before the FIRST tool call of any session / any turn that will involve a tool call counts' framing to narrower triggers so the skill does not compete with every other skill on every turn.

Move activation-config and session-start-hook mechanics out of the description and into the body or environments.md; the description should state capability, not enforcement wiring.

Add one or two concrete output actions (e.g. 'writes one observation file per finding', 'surfaces a grouped summary at session end') to lift specificity from abstract to enumerated.

Dimension	Reasoning	Score
Specificity	Names the domain and several concrete actions — 'Monitors task execution', 'Captures patterns, user corrections and methodology worth preserving as reusable skills' — but the actions stay somewhat abstract (patterns, methodology) rather than fully enumerated.	4 / 5
Completeness	Explicitly answers both 'what' (monitors/captures skill-improvement opportunities) and 'when' ('Use during ANY multi-step task…', 'Also triggers in post-task feedback discussions and when the user mentions skill observations, the observation log, or skill taxonomy') with concrete trigger phrases.	5 / 5
Trigger Term Quality	Strong natural-term coverage — 'multi-step task', 'agentic workflow', 'work session', 'post-task feedback', 'skill observations', 'observation log', 'skill taxonomy', and the 'One Skill to Rule Them All' catchphrase — with a few common phrasings still absent.	4 / 5
Distinctiveness Conflict Risk	The directive to 'invoke this skill before the FIRST tool call of any session' and 'any turn that will involve a tool call counts' is very broad and would fire on nearly every session, creating high overlap risk with many other skills despite its niche trigger phrases.	2 / 5
	Total	15 / 20 Passed

Validation

87%

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 — 14 / 16 Passed

Validation for skill structure

Criteria	Description	Result
skill_md_line_count	SKILL.md is long (729 lines); consider splitting into references/ and linking	Warning
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning

	Total	14 / 16 Passed

Reviewed

2 days ago

Table of Contents

Discovery Implementation Validation