fmind/agent-supagents / .agents/skills/harnesses
harnesses
UnverifiedAdd, update, or verify coding-harness targets, scaffolds, examples, and isolated host discovery probes in agent-supagents.
Harness Targets
Maintain target definitions, scaffolds, examples, and isolated discovery probes across supported coding harnesses without regressing compilation or safety contracts.
Workflow
- Check upstream host documentation for native paths, file suffixes, frontmatter/TOML fields, tool names, and standalone agent listing flags.
- Update the code and example assets in lockstep:
- src/supagents/config.py:
DEFAULT_TARGETS(global_path,project_path,filename_suffix,format,body_key). - src/supagents/cli.py:
_INIT_TARGETSdefaults (keepAGYat defaultmainAgent: truesoagy agentdiscovers it). - examples/reviewer.md: valid current native metadata for all bundled targets.
- scripts/check_hosts.py: CLI executable and standalone discovery command.
- src/supagents/config.py:
- Update unit tests under tests/ and run the offline host discovery probe:
mise run check:hosts - Sync documentation strictly separating compilation, discovery, and runtime enforcement: docs/compatibility.md, docs/usage.md, README.md, and skills/supagent/SKILL.md.
- Run the full gate:
mise run all
Gotchas
- Isolate host probes: compile global probes (such as
AGY) into a temporaryHOME, pass--standaloneor equivalent non-daemon flags (opencode debug agents --standalone), never execute live model calls, and never print raw captured host stdout/stderr. - No cross-harness translation: keep Supagents as a pass-through metadata + body compiler; do not introduce host schema translators or model aliases.
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| reverse-engineer-anythingmorluto/reaReverse engineer native, managed, Electron/JavaScript, packaged, firmware, and browser targets with REA, and build agent environments (AgentEnv bundles) of web applications from a live instance or a capture, and of Linux desktop apps with a task from a request. Use shipped-artifact or requested runtime evidence to explain features, compare versions, decompile code, or guide a reconstruction. Skip REA for ordinary source-repository architecture analysis. | Stars: 467k | Updated | Claude CodeCodex | Verified | Install |
| diagnosing-superpowersobra/superpowersUse when a superpowers session went wrong and your human partner wants to know why — repeated work, ignored plans, stumbles, poor results, a skill that didn't fire, "it took too long", "why is it so expensive", "what is it doing" — or wants to build a bug report for the superpowers maintainers, for the current session or a past one identified by id or path, on any harness. | Stars: 319k | Updated | Claude CodeCodex | Verified | Install |
| dispatching-parallel-agentsobra/superpowersUse when facing 2+ independent tasks that can be worked on without shared state or sequential dependencies | Stars: 319k | Updated | Claude CodeCodex | Verified | Install |
| executing-plansobra/superpowersUse when executing an implementation plan in the current session as the implementer yourself — your human partner chose inline execution, or no subagent tool is available | Stars: 319k | Updated | Claude CodeCodex | Verified | Install |