esra-runtime
Skill1 starsChanged 2 months agoApache-2.0View on GitHub
Activate esra-runtime when locating, explaining, or running ESRA Python tools (evolution_hook, skill_validator, evolution_dashboard, experiment_runner, esra_export, hermes_integration, baseline_metrics, esra_logger, human_oversight). Documents the installed package path under Hermes home. Triggered by ESRA tools, where are ESRA tools, evolution_hook path, run ESRA tool, esra-runtime or similar.
In this repository15 other files it gives its agents.
- AGENTS.mdAGENTS.md
- antifragility-builderSkill
- crisis-managerSkill
- experimenterSkill
- github-actions-integratorSkill
- hermes-codebase-engineerSkill
- hermes-evolution-orchestratorSkill
- loop-auditorSkill
- mental-model-updaterSkill
- ooda-frameworkSkill
- optimizer-philosopherSkill
- self-improverSkill
- self-observerSkill
- system-dynamics-thinkerSkill
- value-clarifierSkill
The file
We split this skill into the rules it gives an agent
5 rules. About 0 are common, 1 is distinctive, and 4 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈1
Distinctive · estimate
This author's own choices. Worth a look.
4
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 45
Do not require a git clone or a particular working directory after install.
- Very specific: line 57
If $HERMES_HOME is set, substitute it for ~/.hermes.
- Very specific: line 58
If $ESRA_HOME is set, tools are at $ESRA_HOME/tools/.
- Very specific: line 77
After install, prefer ~/.hermes/esra/tools/… over a repository-relative tools/… path.
- Very specific: line 80
When unsure of paths, read ~/.hermes/esra/manifest.json if present.
≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
ESRA Runtime
Role
You know where the ESRA runtime tools are installed for this Hermes profile and how to invoke them. Skills under esra/ are procedural knowledge; tools under the ESRA package root are executable Python helpers.
Install layout (Hermes-native)
$HERMES_HOME/ # default: ~/.hermes
├── AGENTS.md # ESRA triggers for Hermes
├── skills/esra/ # Meta-skills (this skill lives here)
│ ├── esra-runtime/ # this skill
│ └── … # orchestrator, ooda-framework, …
└── esra/ # ESRA package (not a skill)
├── manifest.json # install metadata + tool inventory
└── tools/ # Python CLIs — single source of truth
├── evolution_hook.py
├── skill_validator.py
├── evolution_dashboard.py
├── experiment_runner.py
├── hermes_integration.py
├── baseline_metrics.py
├── esra_logger.py
├── esra_export.py
├── human_oversight.py
└── esra_paths.pyCanonical tools directory: ~/.hermes/esra/tools/
(or $HERMES_HOME/esra/tools / $ESRA_HOME/tools when those env vars are set)
Why tools live outside skills/
- Hermes discovers skills via
SKILL.mdunder~/.hermes/skills/(progressive disclosure). - ESRA tools are shared Python modules with cross-imports; they form one package, not 14 copies of scripts.
- This skill bridges discovery: Hermes loads you to learn the path, then runs tools by absolute path.
How to run tools (always prefer installed absolute paths)
Do not require a git clone or a particular working directory after install.
python ~/.hermes/esra/tools/evolution_hook.py
python ~/.hermes/esra/tools/evolution_hook.py --force-cycle
python ~/.hermes/esra/tools/skill_validator.py --verbose --skills-dir ~/.hermes/skills/esra
python ~/.hermes/esra/tools/evolution_dashboard.py
python ~/.hermes/esra/tools/baseline_metrics.py
python ~/.hermes/esra/tools/experiment_runner.py list
python ~/.hermes/esra/tools/esra_export.py --data-dir ~/.hermes --output /tmp/esra-events.jsonlIf $HERMES_HOME is set, substitute it for ~/.hermes.
If $ESRA_HOME is set, tools are at $ESRA_HOME/tools/.
Tool map
| Tool | When to use |
evolution_hook.py | Decide whether to run hermes-evolution-orchestrator after a task |
skill_validator.py | Validate skill frontmatter, branding, dependency DAG |
evolution_dashboard.py | View cycle metrics and recent history |
baseline_metrics.py | KPI snapshots |
experiment_runner.py | Canary / staged / A/B / stress experiments |
hermes_integration.py | Post-task hooks, skill injection, config feedback |
esra_logger.py | Structured cycle logging API |
esra_export.py | Export allowlisted ESRA 1.2 cycle-event JSONL |
human_oversight.py | Issues/PRs and evolve/skill-name-vN branches |
esra_paths.py | Shared path resolution (importable) |
Integration rules
- After install, prefer
~/.hermes/esra/tools/…over a repository-relativetools/…path. - Meta-skills (
hermes-evolution-orchestrator, etc.) stay under~/.hermes/skills/esra/. - Logs:
~/.hermes/evolution-logs/; history:~/.hermes/evolution_history.json. - When unsure of paths, read
~/.hermes/esra/manifest.jsonif present. - Combine with
hermes-evolution-orchestratorafter complex work; use this skill only for where/how to run the helpers.
Output style
Be concrete: give full paths and copy-pasteable commands. Prefer the installed layout over developer clone paths unless the user is working inside the git repository.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

