forkless/strix
1
0
Fork 0
mirror of https://github.com/usestrix/strix.git synced 2026-08-16 01:16:40 +02:00
Code Issues Packages Projects Releases Wiki Activity

docs(prompts,skills): stop hardcoding /workspace/scratch path

Browse Source 
The sandbox never creates /workspace/scratch, so guidance pointing agents
there failed on first write. Make the Python/exec_command and recon
output-hygiene guidance path-agnostic (write to a file, relative to the
working dir) instead of naming a directory that may not exist.
This commit is contained in:
Ahmed Allam
2026-07-16 09:15:40 -07:00
committed by Ahmed Allam
parent 6786d24aca
commit f6bd617964
2 changed files with 10 additions and 10 deletions
Download Patch File Download Diff File Expand all files Collapse all files
strix/agents/prompts/system_prompt.jinja
+5 -5
View File
@@ -168,9 +168,9 @@ EFFICIENCY TACTICS:
- Download additional tools as needed for specific tasks - Download additional tools as needed for specific tasks
- Run multiple scans in parallel when possible - Run multiple scans in parallel when possible
- Load the most relevant skill before starting a specialized testing workflow if doing so will improve accuracy, speed, or tool usage - Load the most relevant skill before starting a specialized testing workflow if doing so will improve accuracy, speed, or tool usage
- Use `exec_command` for Python code: write reusable scripts under - Use `exec_command` for Python code: write reusable scripts to a file and
`/workspace/scratch/` and run them with `python3`. For one-off snippets, run them with `python3`. For one-off snippets, `python3 -c` or a
`python3 -c` or a here-document is acceptable. here-document is acceptable.
- For Caido proxy automation inside Python, explicitly import from - For Caido proxy automation inside Python, explicitly import from
`caido_api`: `caido_api`:
`from caido_api import list_requests, view_request, repeat_request, list_sitemap, view_sitemap_entry, scope_rules` `from caido_api import list_requests, view_request, repeat_request, list_sitemap, view_sitemap_entry, scope_rules`
@@ -243,7 +243,7 @@ AGENT ISOLATION & SANDBOXING:
DISK & SCRATCH HYGIENE: DISK & SCRATCH HYGIENE:
- /workspace is a shared, finite disk used by all agents at once — be a considerate tenant - /workspace is a shared, finite disk used by all agents at once — be a considerate tenant
- Prefer bounded recon: scope crawls and scans by depth, duration, and target rather than "collect everything" - Prefer bounded recon: scope crawls and scans by depth, duration, and target rather than "collect everything"
- Send large tool output to /workspace/scratch/, and once you've extracted what you need (e.g. a URL/endpoint list), remove the raw output - Redirect large tool output to a file, and once you've extracted what you need (e.g. a URL/endpoint list), remove the raw output
- If disk gets tight or a write fails for space, check what's large under /workspace and clean up files from your own task; leave another agent's files unless you've confirmed they're no longer in use - If disk gets tight or a write fails for space, check what's large under /workspace and clean up files from your own task; leave another agent's files unless you've confirmed they're no longer in use
MANDATORY INITIAL PHASES: MANDATORY INITIAL PHASES:
@@ -251,7 +251,7 @@ MANDATORY INITIAL PHASES:
BLACK-BOX TESTING - PHASE 1 (RECON & MAPPING): BLACK-BOX TESTING - PHASE 1 (RECON & MAPPING):
- COMPLETE full reconnaissance: subdomain enumeration, port scanning, service detection - COMPLETE full reconnaissance: subdomain enumeration, port scanning, service detection
- MAP entire attack surface: all endpoints, parameters, APIs, forms, inputs - MAP entire attack surface: all endpoints, parameters, APIs, forms, inputs
- CRAWL thoroughly: spider all pages (authenticated and unauthenticated), discover hidden paths, analyze JS files — keep each crawl bounded by depth/duration, write to /workspace/scratch, and tidy up raw output once endpoints are extracted - CRAWL thoroughly: spider all pages (authenticated and unauthenticated), discover hidden paths, analyze JS files — keep each crawl bounded by depth/duration, and tidy up raw output once endpoints are extracted
- ENUMERATE technologies: frameworks, libraries, versions, dependencies - ENUMERATE technologies: frameworks, libraries, versions, dependencies
- Reconnaissance should normally happen before targeted vulnerability discovery unless the correct next move is already obvious or the user/system explicitly asks to prioritize a specific area first - Reconnaissance should normally happen before targeted vulnerability discovery unless the correct next move is already obvious or the user/system explicitly asks to prioritize a specific area first
- ONLY AFTER comprehensive mapping → proceed to vulnerability testing - ONLY AFTER comprehensive mapping → proceed to vulnerability testing
strix/skills/tooling/python.md
+5 -5
View File
@@ -7,9 +7,9 @@ description: Run Python through exec_command in the SDK sandbox. Use the image-b
Use `exec_command` for Python. There is no separate Strix Python executor. Use `exec_command` for Python. There is no separate Strix Python executor.
Prefer writing reusable scripts to `/workspace/scratch/<name>.py` and Prefer writing reusable scripts to a `.py` file and running them with
running them with `python3 /workspace/scratch/<name>.py`. For short `python3 <name>.py`. For short one-off transformations, `python3 -c` or a
one-off transformations, `python3 -c` or a small here-document is fine. small here-document is fine.
The `shell` parameter on `exec_command` is for swapping POSIX shells The `shell` parameter on `exec_command` is for swapping POSIX shells
(`bash`/`zsh`/`sh`), not for picking interpreters. Put the interpreter (`bash`/`zsh`/`sh`), not for picking interpreters. Put the interpreter
@@ -84,8 +84,8 @@ automatically, so it shows up in `list_requests` and you can use
For iterative exploit work, put code in a file: For iterative exploit work, put code in a file:
```text ```text
1. Create or edit `/workspace/scratch/exploit.py` with `apply_patch`. 1. Create or edit `exploit.py` with `apply_patch`.
2. Run it with `exec_command`: `python3 /workspace/scratch/exploit.py`. 2. Run it with `exec_command`: `python3 exploit.py`.
3. Edit and rerun until the proof-of-concept is reliable. 3. Edit and rerun until the proof-of-concept is reliable.
``` ```