CtrlK
BlogDocsLog inGet started
Tessl Logo

pantheon-ai/save-context

Checkpoint the current session to .context/session/CONTEXT-llm.md with a structured LLM-optimised summary.

58

Quality

73%

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

Overview
Quality
Evals
Security
Files

Quality

Content

50%Weight 40%Scale 1-3

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-organized into clear phases with concrete commands and useful guardrails, but it relies on bundle files (reference.md for the CONTEXT template, scripts/upsert-index.sh) that are absent, undermining both actionability and progressive disclosure.

Suggestions

Add the missing reference.md (CONTEXT file template, status mapping, INDEX.md upsert logic) and scripts/upsert-index.sh to the bundle, or inline the template so the skill is executable from SKILL.md alone.

Fold the 'read existing context file before overwriting' guard into the Phase 1/3 workflow as an explicit checkpoint, and add a post-write verification step (e.g. confirm the file exists and INDEX.md was upserted).

Trim the editorializing Philosophy section and merge 'When Not to Use' with 'Anti-Patterns' to remove redundancy and tighten token use.

DimensionReasoningScore

Conciseness

The body is dense and avoids explaining basics Claude knows, but the Philosophy section is editorializing and 'When Not to Use' overlaps with 'Anti-Patterns', so it could be tightened; not a 1 because no padded concept explanations are present.

2 / 3

Actionability

Concrete commands are given (rtk ls, mkdir -p ... && mv ...) and Phase 2 specifies sections with token budgets, but the core output — the CONTEXT file template — is deferred to 'reference.md' which is not present in the bundle, leaving a key detail missing.

2 / 3

Workflow Clarity

The three phases are clearly sequenced with stream-regex validation and status-gated archival, but the 'read existing file before overwrite' guard lives only in Anti-Patterns rather than the workflow, and there is no post-write verification; the destructive-ops guideline caps this at 2.

2 / 3

Progressive Disclosure

A single one-level-deep reference is clearly signaled ('See reference.md for ...'), but the referenced file and scripts/upsert-index.sh do not exist in the bundle, so the navigation is broken rather than achieving easy navigation.

2 / 3

Total

8

/

12

Passed

Description

90%Weight 40%Scale 1-3

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, third-person description that explicitly states what the skill does and when to use it, with a rich set of natural trigger phrases. Its only weakness is specificity: it names a single saving action rather than enumerating multiple concrete capabilities.

DimensionReasoningScore

Specificity

Names a concrete action and target ('Save session to CONTEXT-llm.md with conversation summary') but lists only one main action rather than multiple specific concrete actions, matching the 'names domain and some actions, but not comprehensive' anchor.

2 / 3

Completeness

Clearly answers both what ('Save session to CONTEXT-llm.md with conversation summary') and when ('Use when saving work, checkpointing progress, preserving session state') with explicit triggers, matching the top anchor.

3 / 3

Trigger Term Quality

Explicit natural trigger phrases users would actually say ('save context', 'save session', 'checkpoint', 'save my progress') give good coverage; not the level below because both common variations and explicit quoted triggers are present.

3 / 3

Distinctiveness Conflict Risk

The CONTEXT-llm.md file-naming niche plus context-specific triggers make it clearly distinguishable and unlikely to fire for the wrong skill.

3 / 3

Total

11

/

12

Passed

Validation

75%

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

Validation12 / 16 Passed

Validation for skill structure

CriteriaDescriptionResult

allowed_tools_field

'allowed-tools' contains unusual tool name(s)

Warning

frontmatter_unknown_keys

Unknown frontmatter key(s) found; consider removing or moving to metadata

Warning

relative_links

Relative link issues: 1 missing

Warning

referenced_paths_exist

Referenced path issues: 1 missing

Warning

Total

12

/

16

Passed

Reviewed

Table of Contents