All skills
oaustegard avatar

/iterating

@5e58100

Multi-conversation methodology for iterative stateful work with context accumulation. Use when users request work that spans multiple sessions (research, debugging, refactoring, feature development), need to build on past progress, explicitly mention iterative work, work logs, project knowledge, or cross-conversation learning.

Use this Skill: https://skilld.dev/gh/oaustegard/claude-skills/iterating

This session only. Nothing lands on disk.

referenceschat-environment.md

≈226 tokens on demand. Your agent reads this file only when SKILL.md points to it.

Chat Environment (Claude.ai)

Claude.ai web/chat/native app - no direct file system access.

Persistence

Save WorkLog to outputs for download:

cp worklog.md /mnt/user-data/outputs/[Project]-WorkLog-vN.md

Provide download link:

[Download WorkLog vN](computer:///mnt/user-data/outputs/[Project]-WorkLog-vN.md)

Retrieval

From uploaded file:

  • User uploads WorkLog file
  • Located in /mnt/user-data/uploads/
  • Find: ls -1 /mnt/user-data/uploads/*WorkLog*.md | sort -V | tail -1

From pasted content:

  • User pastes WorkLog content directly
  • Detect YAML frontmatter with version: and status:
  • Parse and continue

From Project Knowledge:

  • User-curated files mounted at /mnt/project/
  • RAG system may also inject relevant content into context
  • Check: ls /mnt/project/*WorkLog*.md 2>/dev/null
  • Read directly if present, or rely on RAG-injected context

Source: SKILL.md on GitHub

1 warning4mo4 checks · Risk SAFE
  • Gen Agent Trust Hub4mo

    The skill establishes a robust methodology for stateful, iterative work using 'Work Logs' to maintain context across multiple sessions. While functionally safe and containing no malicious code, the skill's reliance on reading and executing 'HIGH priority' tasks from external, potentially user-controlled files creates a surface for indirect prompt injection.

  • Socket4mo

    No alerts

  • Snyk4mo

    Risk: LOW · No issues

  • Runlayer7mo

    10/10 files flagged

Signed by skilld at 5e58100. This ties the file your Agent reads to that commit on GitHub. It does not review the instructions.

Last checked against GitHub yesterday.

Activeupdated 3 weeks ago
metadata
{
  "version": "1.2.0"
}

README badge

README badge for oaustegard/claude-skills/iterating