Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx agentmods add commands/gobing-ai/superskill/absorb-codegit clone --depth 1 https://github.com/gobing-ai/superskillWhat it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5 | $0.00012 | $0.00795 |
| Opus 5 | $0.00006 | $0.00398 |
| Sonnet 5 | $0.00002 | $0.00159 |
| Haiku 4.5 | $0.00001 | $0.00080 |
Grade B, and why
absorb-code scanned grade B with 1 finding against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured yesterday.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Enumerates other installed skillsmediumAgent snooping
Other skills' SKILL.md files reveal prompts, capabilities and secrets that should be invisible to peers.
- Verify the symlink is intact: `ls -la .agents/skills` should point to `.claude/skills`. How it starts
The opening of the file, as written. The whole thing — 80 lines — stays where its author put it; the contents beside it link to each section on GitHub.
You are finishing a convergence absorption cycle. converge apply has just copied files from a source project into .claude/skills/, .claude/commands/, or other destination paths under this project.
ARGUMENTS supplies the source project path (--from), a description of what was absorbed and any special instructions (--description), and optionally the review artifact path. Read $ARGUMENTS to extract these values.
0. Understand what happened
Parse the arguments:
--from <source-project>— the source project the convergence scanned. Required.--description <text>— what was absorbed (skills, commands, configs, code), why, and any cleanup preferences or constraints. Required.[review-artifact]— path to the review JSON fromconverge review. Optional; if omitted, infer from the most recent file indocs/reviews/.
If any required argument is missing, ask for it before proceeding.
1. Review what was absorbed
- Run
git status -sto see every file the apply step created or modified. - If a review artifact was provided or found, read it to understand the classification decisions.
- For each new or changed file, check whether it contains project-specific references that should have been classified as
project-specificorsensitive:- Absolute paths pointing to the source project's directory
- Organization names, cloud account IDs, or deployment targets
- Hardcoded secrets, tokens, or internal endpoints
- References to source-project-only packages or tooling
2. Clean up drift
For every file flagged above:
- Replace source-project names/paths with the current project's equivalents.
- Strip or generalize deployment targets, account IDs, and cloud-specific config.
- If a file is too entangled with the source project to generalize, classify it as blocked and remove it. Do NOT leave half-cleaned files in the tree.
Apply any additional instructions from --description.
3. Verify capability registration
- Every new
.claude/skills/<name>/SKILL.mdmust be discoverable. Verify the skill directory exists with aSKILL.mdand that its frontmatter is valid. - Every new
.claude/commands/<name>.mdmust follow the project's command format. - If an absorbed capability is mode-scoped, confirm it carries a
supported-modesfrontmatter annotation listing the modes it supports.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- yesterday First seen · 80 lines · 0 tokens per session scan B 273a5c647694
absorb-code is a command published in the GitHub repository gobing-ai/superskill (5 stars, last pushed 6d ago), licensed Apache-2.0. It adds 12 tokens to every session and 795 once invoked, about $0.0001 per session on Opus 5. A static security scan graded it B with 1 finding (enumerates other installed skills). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other commands, from other repositories
checklist
Generate a custom checklist for the current feature based on user requirements.
clarify
Identify underspecified areas in the current feature spec by asking up to 5 highly targeted clarification questions and encoding answers back into the spec.
specify
Create or update the feature specification from a natural language feature description.
analyze
Perform a non-destructive cross-artifact consistency and quality analysis across spec.md, plan.md, and tasks.md after task generation.
converge
Assess the current codebase against the feature's spec, plan, and tasks, then append any remaining unbuilt work as new tasks to tasks.md so implement can complete it.
implement
Execute the implementation plan by processing and executing all tasks defined in tasks.md.