openai/boltz-small-molecule-screen

Workflow

If boltz-api is missing from PATH, use boltz-cli-setup for install/update guidance before retrying. If a command reports missing or expired authentication, use boltz-cli-setup to start boltz-api auth login --device-code before retrying; do not ask permission first. If the agent host sandbox blocks boltz-api install/auth/API calls, use boltz-cli-setup to request the host sandbox bypass/escalation needed for user-wide CLI install, browser login, credential storage, temp files, or API access before retrying.

Use this skill when the user already has candidate molecules.

1. Normalize the library from raw SMILES, a CSV (auto-detect the SMILES column), .smi, or .txt into the molecules list. Each entry is {smiles, id?}; the optional id is echoed back as external_id on each result.
2. Normalize the target: one or more protein sequences into target.entities, plus optional pocket_residues (0-based) and/or reference_ligands (SMILES of known binders to help locate the pocket).
3. Keep default server-side filtering unless the user asks for custom filters — only add molecule_filters on explicit request.
4. Author the payload YAML or JSON, run estimate-cost, show the USD cost, wait for explicit confirmation.
5. start to submit (synchronous). Capture the ID.
6. Launch download-results with the agent runtime's background/non-blocking command facility — it polls, paginates list-results, downloads every per-hit structure, and exits when terminal. In Claude Code, use Bash with run_in_background: true. In Codex, run download-results as a foreground shell command with yield_time_ms: 1000; if Codex returns a session_id, keep it for optional same-thread polling, but treat download-status plus the run directory as the durable source of truth. In Codex app/desktop runtimes that expose same-thread heartbeat automations, create a heartbeat that checks download-status periodically and posts a concise completion or failure update when the download reaches a terminal state. After launching the downloader, always report the job ID, run name, and output directory. Include the next check cadence if the heartbeat was created; otherwise include the download-status command.
7. When done, rank from <output-root>/<run-name>/results/index.jsonl. Sort by binding_confidence for hit discovery or optimization_score for lead optimization; these are parallel intents, not a fallback hierarchy. Report the top 5-10 hits with smiles, the chosen ranking metric, key confidence metrics, and structure path. Each result also carries a free adme block (solubility, permeability, lipophilicity) — include it for developability triage when the user cares about ADME, or when a top hit looks risky. Read references/results.md for output layout, metrics, ADME, and filtered-input accounting.

Command Pattern

# Replace placeholders with concrete absolute paths before running.
# Use a short descriptive run name, for example: sm-screen-<target>-<library>-v1

boltz-api small-molecule:library-screen estimate-cost \
  --input @yaml:///absolute/path/payload.yaml

boltz-api small-molecule:library-screen start \
       --idempotency-key "<run-name>" \
       --input @yaml:///absolute/path/payload.yaml \
       --raw-output --transform id

# Copy the printed job ID into this command, then launch it in the agent
# runtime's background/non-blocking mode.
# Claude Code: Bash with run_in_background=true.
# Codex: foreground shell command with yield_time_ms=1000; keep the returned session_id if one is provided.
# Do not append "&" or use nohup in Codex.
boltz-api download-results \
  --id "<job-id-from-start>" --name "<run-name>" \
  --root-dir "/absolute/path/boltz-experiments" \
  --poll-interval-seconds 30
# -> /absolute/path/boltz-experiments/<run-name>/results/<pres_*>/...

Payload keys are molecules, target, molecule_filters — the API body field names, not the direct CLI flag names --molecule / --target / --molecule-filters.

Always Do This

• Keep payload field names exactly as the API body names shown in references/api.md.
• Use absolute paths for the output root, payload files, and embedded target files. Do not cd into the run directory for follow-up commands; pass the same --root-dir and use absolute paths so later relative paths do not drift.
• Prefer one merged top-level payload via --input @yaml:///absolute/path/payload.yaml or @json:///absolute/path/payload.json for estimate-cost and start. Keep --idempotency-key and --workspace-id top-level; if they also appear inside --input, the top-level flags win.
• Direct object flags still work as overrides, such as --target @yaml:///absolute/path/target.yaml, --molecule-filters @json:///absolute/path/filters.json, or repeated --molecule @json:///absolute/path/mol-1.json entries. Piped YAML / JSON on stdin also works, but it must use API body field names. Never use @file:// or @./.
• Treat pocket residue indices as 0-based.
• Do not invent medicinal-chemistry filters. Only add molecule_filters if the user asks; mention the catalog as an option.
• Use the same slug as both --idempotency-key at submit and --name on download-results so re-runs resume via .boltz-run.json.
• In permission-gated agents such as Claude Code, keep each Boltz call as a top-level command that starts with boltz-api. Prefer concrete arguments over sh -c, inline environment assignments, aliases, wrapper scripts, loops, or pipelines around the boltz-api invocation unless the user already allowed that exact command form. Use --raw-output --transform id, read the printed ID, then paste that literal ID into the next download-results command.
• Prefer the agent runtime's background/non-blocking command mode for download-results. In Codex specifically, keep download-results in the foreground and set the shell tool yield to 1000 ms; Codex will return a session_id if the command is still running. Do not append & or use nohup in Codex because the tool runner may clean up shell-backgrounded descendants before .boltz-run.json is fully written.
• After the background/session starts, do not manually wait on it or run ad hoc polling loops. download-results emits JSONL progress on stderr by default; add --progress-format text --verbose only when you explicitly want human-readable logs.
• In Codex app/desktop runtimes with same-thread heartbeat automation support, schedule a heartbeat after launching download-results. The heartbeat should run boltz-api --format json download-status --name "<run-name>" --root-dir "/absolute/path/boltz-experiments" and stop once terminal. Choose cadence by molecule count: under 100 -> every 1-2 minutes; 100-1,000 -> every 5 minutes; over 1,000 -> every 15 minutes. Post only material status changes or terminal completion/failure. Poll the saved session_id with an empty write_stdin only for interactive, user-requested progress checks. Never run a manual poll loop in the current turn.
• If the current host has no heartbeat automation support, do not claim an automatic next check. Report the job ID, run name, output directory, and the command needed to check download-status.
• If detached download needs to be restarted, re-run boltz-api download-results with the same --name "<run-name>" and the same --root-dir.
• Cost is a flat $0.025 per molecule (size-independent). estimate-cost returns the authoritative total — always use it.
• Poll interval: --poll-interval-seconds 30 is a reasonable downloader default. Wall-clock time scales roughly with the number of molecules: under 100 often finishes in a few minutes, 100-1,000 may take several minutes to tens of minutes, and larger screens can take longer or hours depending on inputs and system load. Don't quote a fixed duration, and never tell the user a 10-candidate screen will take 30 minutes or hours.

Escape Hatch

• Payload reference: https://api.boltz.bio/docs/api/python/resources/small_molecule/subresources/library_screen/methods/start
• CLI flag names: boltz-api small-molecule:library-screen start --help

Read references/api.md for the molecules, target, and molecule_filters shapes, including the built-in SMARTS filters and RDKit descriptor ranges. Read references/results.md after download when ranking hits or explaining missing/filtered inputs.

Outputs

Rank from results/index.jsonl after download-results; use references/results.md for the local file layout, metric meanings, and filtered-input accounting.