Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx agentmods add skills/fl-sean03/opensdl/develop-workflownpx skills add fl-sean03/OpenSDL --skill develop-workflowgit clone --depth 1 https://github.com/fl-sean03/OpenSDLWhat it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5 | $0.00035 | $0.00372 |
| Opus 5 | $0.00017 | $0.00186 |
| Sonnet 5 | $0.00007 | $0.00074 |
| Haiku 4.5 | $0.00003 | $0.00037 |
Grade A, and why
develop-workflow scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured yesterday.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
What it actually says
Develop a simulated workflow
Inputs
- simulation manifest path
- workflow path
- intended inputs, outputs, and failure behavior
- representative JSON inputs
Procedure
- Read the manifest, capability contracts, and a nearby tested workflow.
- Confirm that
spec.environmentissimulation, review every enabled adapter, and trace every selected capability through its enabled binding to the intended adapter plugin. - Define typed inputs, step dependencies, value bindings, resources, timeouts, retries, and outputs.
- Add or update focused workflow validation and execution tests.
- Run
uv run --locked opensdl validate MANIFEST --workflow WORKFLOW. - Run
.agents/skills/develop-workflow/run.sh WORKFLOW MANIFEST 'INPUTS_JSON'. Before the runtime starts any adapter, the helper permits only the reviewed OpenSDL referencesimulated-lab,local-compute, andhuman-taskentry points and verifies every enabled plugin's module and distribution provenance. Qualify and review any addition before changing that allowlist. - Inspect failure events when the run does not complete. Use
debug-runfor a persisted failure. - Export the run when the user needs portable execution evidence.
Completion
The workflow validates, its tests pass, and a representative simulation run produces the expected outputs and persisted evidence.
Stop conditions
Stop before execution when the manifest environment is not simulation, a binding is missing or
ambiguous, any enabled plugin is outside the reviewed simulation allowlist, or the workflow needs
an operation that no typed capability provides.
What ships with it
1 file beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- yesterday First seen · 40 lines · 35 tokens per session scan A 5936c44a7e10
develop-workflow is a skill published in the GitHub repository fl-sean03/OpenSDL (102 stars, last pushed yesterday), licensed Apache-2.0. It adds 35 tokens to every session and 372 once invoked, about $0.0002 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other skills, from other repositories
scitex
Skill "scitex" from scitex-ai/scitex-python, covering scitex — umbrella package, mechanics, sub-skills index, quick reference and mandatory leaves.
stx.cli
Command-line interface for SciTeX platform operations — 30+ subcommands with lazy loading. Specialization of scitex-cli-convention.
jupyter-notebook
Jupyter notebook structure, reproducibility, and best practices for data science.
numpy-docs
NumPy 2.5 — array creation, data types, indexing, broadcasting, ufuncs, structured arrays, I/O, C-API.
rrdoctor-verify
Use rrdoctor as the deterministic, offline definition of done for preparing a research repository for Artifact Evaluation or public release.
seamless-remote-debugging
Guides AI agents through debugging failures in Seamless remote execution (hashserver, database, jobserver, daskserver). Covers the multi-service topology, error propagation across repos, finding server-side logs and PID files, restarting services cleanly, and clearing caches to prevent false passes. Triggers when a…