The Agent Skills Directory

[COMMAND_EXECUTION]: The utility script render-graphs.js executes the system command dot from the Graphviz suite using child_process.execSync. It processes text extracted from markdown code blocks and writes generated SVG files to the local filesystem.
[PROMPT_INJECTION]: The skill documentation, specifically SKILL.md and persuasion-principles.md, instructs the use of directive language such as "YOU MUST", "Never", and "No exceptions". These are used to implement an "Authority" principle designed to override agent rationalization and enforce compliance with documentation standards.
[PROMPT_INJECTION]: The skill methodology involves processing agent-generated feedback ("rationalizations") and untrusted data ("pressure scenarios") to refine instructions. This creates an indirect surface where external data could influence agent behavior, though it is mitigated by the skill's focus on manual documentation hardening.
[DATA_EXFILTRATION]: No sensitive file access or network transmission to external or unknown domains was detected.
[EXTERNAL_DOWNLOADS]: Documentation within the skill references external libraries like pdfplumber and pypdf as implementation examples but does not automate their download or execution from remote sources.

writing-skills