Name: context-surfing
Rating: 63.2 (1 reviews)
Author: pskoett

context-surfing

Monitors context window health during large, long-running, multi-session, or explicitly context-sensitive work. Uses bounded external anchors, one cold-context review when self-recovery is uncertain, and a clean handoff on real drift. Do not auto-activate merely because a medium task has an intent frame and plan.

Quality

75%

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

Fix and improve this skill with Tessl

tessl review fix ./plugin/skills/context-surfing/SKILL.md

The canonical home for this skill is context-surfing in pskoett/pskoett-ai-skills

Quality

Content

63%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 content is a thorough, well-structured monitoring protocol with concrete commands and clear multi-step workflows, but it over-relies on extended metaphor and restated procedures, and inlines material that would benefit from being split into reference files.

Suggestions

Trim the surf metaphor to a single short Mental Model block and remove restated protocols — describe the pre-commit anchor check once and cross-reference it from Riding the Wave instead of repeating it.

Move the full handoff template and the interoperability/pipeline-position sections into separate reference files (e.g. references/handoff-template.md, references/pipeline.md) and link to them, keeping SKILL.md a lean overview.

Replace or soften reliance on external artifacts (plan-interview, intent-framed-agent, context-monitor subagent) with self-contained fallbacks, or clearly gate them as optional so the skill is executable standalone.

Dimension	Reasoning	Score
Conciseness	The body is ~420 lines with extended surf metaphor scaffolding ("Paddling in", "the shoulder", "No wipeouts. No zombie sessions.") and restated protocols — the pre-commit anchor check is described in the Monitoring Paradox section and again in Riding the Wave — adding tokens beyond what Claude needs.	3 / 5
Actionability	Provides concrete commands (`entire status`, `entire explain --session <id>`, `ls .context-surfing/handoff-*.md`), a full handoff template, and hook JSON config, though it leans on external skills/subagents (plan-interview, context-monitor) whose availability is uncertain.	4 / 5
Workflow Clarity	Recovery Protocol and Exit Protocol are clearly sequenced with explicit branches, re-anchor limits, and a precedence rule; this monitoring skill is not a destructive/batch operation so the destructive-cap does not apply, leaving only minor checkpoint gaps.	4 / 5
Progressive Disclosure	Structure is present via clear headers and one real bundle script (scripts/handoff-checker.sh), but the large inline handoff template and the lifecycle/interoperability sections could live in separate files, and plugin-bundle references (plugin/agents/context-monitor.md) point outside the skill bundle.	3 / 5
	Total	14 / 20 Passed

Description

87%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 strong: it states concrete capabilities, explicit activation conditions, and a clear boundary against over-triggering. It distinguishes itself well from sibling skills and uses natural trigger language.

Dimension	Reasoning	Score
Specificity	Lists several concrete actions — "Monitors context window health", "Uses bounded external anchors", "one cold-context review", "a clean handoff on real drift" — with only minor coverage gaps, fitting the 'several specific actions' anchor above the 3.	4 / 5
Completeness	Explicitly answers both what ("Monitors context window health... Uses bounded external anchors, one cold-context review... and a clean handoff on real drift") and when ("during large, long-running, multi-session, or explicitly context-sensitive work") with concrete trigger phrases.	5 / 5
Trigger Term Quality	Natural task-shape terms a user would say — "large", "long-running", "multi-session", "context-sensitive", "drift" — give good keyword coverage, though a few synonyms are missing, so it sits above the 3 rather than at 5.	4 / 5
Distinctiveness Conflict Risk	Occupies a clear niche (context-window health monitoring) and actively reduces conflict risk with the explicit boundary "Do not auto-activate merely because a medium task has an intent frame and plan".	5 / 5
	Total	18 / 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
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning

	Total	15 / 16 Passed

Repository: pskoett/pskoett-ai-skills
Path: plugin/skills/context-surfing/SKILL.md
Commit: 8a71d70

Reviewed: 11 days ago

Table of Contents

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.