Skip to content

Sync upstream Strix and vendor Anthropic Cybersecurity Skills v1.2.0#7

Draft
Copilot wants to merge 10 commits intomainfrom
copilot/integrate-all-skills-sync-strix
Draft

Sync upstream Strix and vendor Anthropic Cybersecurity Skills v1.2.0#7
Copilot wants to merge 10 commits intomainfrom
copilot/integrate-all-skills-sync-strix

Conversation

Copy link
Copy Markdown

Copilot AI commented Apr 7, 2026

This updates the branch to the latest upstream usestrix/strix and integrates the Anthropic Cybersecurity Skills v1.2.0 library into Strix. The main change is expanding the custom skill catalog at scale without breaking skill discovery or prompt/tool ergonomics.

  • Upstream sync

    • Fast-forwarded to the latest upstream Strix changes, including the repo’s migration to uv and recent agent / scan-mode updates.
  • Vendored skill catalog

    • Imported the Anthropic Cybersecurity Skills v1.2.0 catalog into strix/skills/custom.
    • Added the full slug-based skill set as native Strix markdown skills so they are loadable through the existing skill pipeline.
  • Catalog scaling changes

    • Refactored skill discovery helpers in strix/skills/__init__.py to centralize category collection and flattening.
    • Changed generated skill descriptions and invalid-skill error messages to summarize the catalog by category and examples instead of enumerating every skill name.
    • Keeps the skill interface usable now that the catalog is hundreds of skills larger.
  • Docs

    • Updated strix/skills/README.md to document the imported Anthropic catalog and the slug-based naming convention used to load these skills.
  • Coverage

    • Added tests for:
      • discovery of imported custom skills
      • loading vendored skills through the existing loader
      • bounded, summary-style descriptions and validation errors for large catalogs

Example of the new load pattern:

create_agent(
    task="Investigate suspicious API traffic",
    name="Traffic Analyst",
    skills="analyzing-api-gateway-access-logs,performing-network-traffic-analysis-with-tshark",
)

0xallam and others added 10 commits March 22, 2026 16:01
Removed specific tool files from Dockerfile and added a directory copy instead.
- Change default model from gpt-5 to gpt-5.4 across docs, tests, and examples
- Remove Strix Router references from docs, quickstart, overview, and README
- Delete models.mdx (Strix Router page) and its nav entry
- Simplify install script to suggest openai/ prefix directly
- Keep strix/ model routing support intact in code

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
- Rewrite system prompt: refusal avoidance, system-verified scope, thorough
  validation mandate, root agent orchestration role, recon-first guidance
- Add authorized targets injection via system_prompt_context in strix_agent
- Add set_system_prompt_context to LLM for dynamic prompt updates
- Prefer python tool over terminal for Python code in tool schemas
- Increase LLM retry backoff cap to 90s
- Replace models.strix.ai footer with strix.ai

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
…odes

Models occasionally output text-only narration ("Planning the
assessment...") without a tool call, which halts the interactive agent
loop since the system interprets no-tool-call as "waiting for user
input." Rewrite both interactive and autonomous prompt sections to make
the tool-call requirement absolute with explicit warnings about the
system halt consequence.
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

5 participants