CtrlK
BlogDocsLog inGet started
Tessl Logo

ac-context-compactor

Manage and compact context for long sessions. Use when context is filling up, creating handoff summaries, optimizing context usage, or preparing for session continuation.

61

Quality

77%

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 ./.claude/skills/ac-context-compactor/SKILL.md

The canonical home for this skill is ac-context-compactor in fernandezbaptiste/Skrillz

SKILL.md
Quality
Evals
Security

Quality

Content

61%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 well-structured and grounded in a real bundle script, with concrete thresholds and a useful summary-format example. Its weaknesses are redundant purpose restatements, an implicit workflow lacking validation before destructive compaction, and a Quick Start that is not fully copy-paste executable.

Suggestions

Delete the Purpose section (or fold it into the opening tagline) — it repeats the frontmatter description nearly verbatim, and merge the abstract Compaction Strategy bullets into the workflow steps.

Add an explicit ordered workflow with a validation checkpoint: after create_compact_summary, verify the summary captures critical state (current feature, active file, in-flight decisions) before discarding context, especially on the above-90% forced handoff path.

Make the Quick Start copy-paste runnable (wrap in an async function, show how project_dir is constructed or show the CLI equivalent from scripts/context_compactor.py) and inline one or two key method signatures in the API Reference.

DimensionReasoningScore

Conciseness

The opening tagline ('Manage context usage and create compact summaries for session continuation') and the Purpose paragraph ('Monitors context usage and creates compact summaries when approaching limits, enabling seamless session continuation...') restate the same sentence twice, and both repeat the frontmatter description. The Compaction Strategy bullets ('Preserve: Critical state and progress') are also abstract filler; the body could be tightened without losing information.

3 / 5

Actionability

The Quick Start shows real API calls (ContextCompactor, should_compact, create_compact_summary — all verified to exist in scripts/context_compactor.py) and the thresholds (85%, 90%) and Summary Structure JSON are concrete. It falls short of anchor 5 because the snippet is not copy-paste runnable as-is (undefined project_dir, no async wrapper) and the API Reference section is just a pointer with no key-method summary.

4 / 5

Workflow Clarity

A sequence exists implicitly (check should_compact -> create summary -> force handoff above 90%), with the threshold tree providing decision points, but there are no explicit ordered steps and no validation checkpoints. Compaction irreversibly discards context, which is a destructive operation, so the missing validation/verification step caps this at 3.

3 / 5

Progressive Disclosure

The body is a clear overview with well-labeled sections, and the implementation detail is appropriately split into scripts/context_compactor.py, referenced one level deep in the API Reference section. Minor gaps: the reference is a plain path rather than a link, and a couple of key method signatures could be inlined before pointing to the full implementation.

4 / 5

Total

14

/

20

Passed

Description

83%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: it explicitly answers both what the skill does and when to use it, in third person, with natural trigger phrases. The main weaknesses are the generic verb 'Manage', a slightly broad 'optimizing context usage' trigger, and missing common synonyms like 'context window' or 'token limit'.

DimensionReasoningScore

Specificity

The description names several concrete actions — 'compact context', 'creating handoff summaries', 'preparing for session continuation' — with minor gaps: 'Manage' is generic filler and monitoring context usage is never explicitly stated. It lists more than the 1-2 actions of the anchor-3 example but is not comprehensive like the anchor-5 example.

4 / 5

Completeness

Both parts are explicitly answered: the what ('Manage and compact context for long sessions') and an explicit 'Use when...' clause with concrete trigger phrases ('context is filling up, creating handoff summaries, optimizing context usage, or preparing for session continuation'). Voice is third person, so no penalty applies.

5 / 5

Trigger Term Quality

Natural phrases like 'context is filling up', 'handoff summaries', and 'session continuation' map well to what a user would actually say. It falls short of anchor 5 because common synonyms such as 'context window', 'token limit', 'compaction', or 'summarize the session' are missing.

4 / 5

Distinctiveness Conflict Risk

Long-session context compaction is a clear niche with distinct triggers, but the trigger 'optimizing context usage' is broad enough to overlap with general session-management or summarization skills. Mostly distinct with minor overlap risk — not the minimal-conflict clarity of anchor 5.

4 / 5

Total

17

/

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
fernandezbaptiste/Skrillz
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.