Tessl • The management layer for AI agent skills

SkillPublic

create-context

Create the baseline session context from the in-folder of raw project documents, classifying each file by priority, writing a manifest, summarising oversized files and producing a token-budgeted baseline. Use when bootstrapping project context or setting up the ctx snapshot once per project. Triggers include "create context", "bootstrap context", "setup context", "init context".

Quality

91%

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

Quality

Content

86%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 a lean, executable workflow with strong guards, explicit anti-patterns justified by WHY, and a clean split of detail into bundled reference files. The remaining gaps are minor: no exact copy command for step 5 and no stated recovery path for validation failures or the destructive --force rerun.

Suggestions

Add the exact copy command for step 5 (e.g., a manifest-driven loop or an include-list copy that preserves subdirectories) so HIGH/MEDIUM copying is copy-paste ready rather than described.

State the recovery loop for step 6: what to do when validate-manifest.sh fails (fix the reported entries, re-run) so validation closes the feedback cycle.

Add a brief check before the destructive --force rerun (e.g., confirm summaries to preserve) since --force discards existing summaries.

Dimension	Reasoning	Score
Conciseness	Every section carries non-obvious, project-specific guidance — the blank-answer AskUserQuestion guard, secret filename patterns, the "tokens are words divided by 0.75" formula, the "one manifest per session" rule — with no filler or explanation of concepts Claude already knows. It is not a 4 because nothing in the body is plainly trimmable.	5 / 5
Actionability	Mostly executable: guard snippets, "./scripts/scan-in-folder.sh .context/session/in", "./scripts/validate-manifest.sh", the sizing table, and BAD/GOOD YAML/bash examples. It is not a 5 because step 5's "Copy HIGH and MEDIUM files into the ctx folder and keep their subdirectories" gives no exact command despite warning against the naive `cp -r`; not a 3 because nearly all guidance is copy-paste ready.	4 / 5
Workflow Clarity	Seven ordered steps with a prerequisite guard ("Stop on a missing in-folder... stop on an existing ctx folder unless --force") and explicit manifest validation, plus a blank-answer recovery loop. It is not a 5 because no recovery path is stated when validate-manifest.sh fails and the destructive "--force discards existing summaries" path has no verification step; not a 3 because validation checkpoints are present, so the batch-operation cap does not apply.	4 / 5
Progressive Disclosure	The body is a workflow overview; schema, sizing rules, baseline template and worked examples are appropriately split into real one-level-deep files (references/reference.md, references/examples.md) listed with one-line descriptions, and the bundled scripts are invoked by exact path. This matches the top anchor.	5 / 5
	Total	18 / 20 Passed

Description

96%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 explicit, third-person, and complete: it states concrete actions for the whole workflow, an explicit 'Use when' clause, and multiple natural trigger phrases. Its only weakness is mild trigger ambiguity against sibling context-management skills.

Dimension	Reasoning	Score
Specificity	The description lists four concrete, comprehensive actions: "classifying each file by priority, writing a manifest, summarising oversized files and producing a token-budgeted baseline" — the full workflow, in third person. It is not a 4 because coverage is complete rather than leaving minor gaps.	5 / 5
Completeness	It explicitly answers both questions: what ("Create the baseline session context from the in-folder of raw project documents...") and when ("Use when bootstrapping project context or setting up the ctx snapshot once per project") with concrete trigger phrases. It matches the top anchor exactly.	5 / 5
Trigger Term Quality	Explicit triggers "create context", "bootstrap context", "setup context", "init context" plus the phrases "bootstrapping project context" and "setting up the ctx snapshot" give comprehensive natural synonym coverage for this niche. It is not a 4 because no commonly used variant of the request is missing.	5 / 5
Distinctiveness Conflict Risk	Niche vocabulary like "ctx snapshot", "in-folder" and "token-budgeted baseline" makes the skill mostly distinct, but generic trigger phrasing such as "setup context" carries minor overlap risk with closely related context skills (save-context/load-context). It is not 5 because of that overlap; not 3 because the scope is clearly narrower than a broadly overlapping skill.	4 / 5
	Total	19 / 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
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

	Total	14 / 16 Passed

Repository: pantheon-org/tekhne
Path: skills/agentic-harness/create-context/SKILL.md
Revision: 9aa89e5

Reviewed: 5 hours ago

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.

create-context

Table of Contents

Is this your skill?