The 170-Line SOUL.md That Made My Hermes Agent Dangerous
Why a single 170-line markdown file — not a secret model or magic framework — is what makes a Hermes Agent push back, hold you accountable, and act like an operator instead of a chatbot.
- Category
- Configuration
- Level
- Community
- Author
- Tony
- Reading time
- 5 min
Why a single 170-line markdown file — not a secret model or magic framework — is what makes a Hermes Agent push back, hold you accountable, and act like an operator instead of a chatbot.
Community flow by Tony. View sourceFlow sections
- The file that changes everything
- Most people train their AI to be useless
- It is required to push back
- It holds you accountable too
- Hermes has a split personality (on purpose)
- It knows exactly what you're building
Section outline mirrored from the public community flow. Use the source page for full prose and examples.
Upstream outline
- The file that changes everything
- Most people train their AI to be useless
- It is required to push back
- It holds you accountable too
- Hermes has a split personality (on purpose)
- It knows exactly what you're building
- The autonomy boundary is brutally simple
- Why "be helpful" doesn't work
- How to build your own SOUL
- Final thought
Section map
The file that changes everything
Frames the file that changes everything for this Configuration workflow, including the operating context to verify in the source page.
Most people train their AI to be useless
Frames most people train their ai to be useless for this Configuration workflow, including the operating context to verify in the source page.
It is required to push back
Frames it is required to push back for this Configuration workflow, including the operating context to verify in the source page.
It holds you accountable too
Frames it holds you accountable too for this Configuration workflow, including the operating context to verify in the source page.
Hermes has a split personality (on purpose)
Frames hermes has a split personality (on purpose) for this Configuration workflow, including the operating context to verify in the source page.
It knows exactly what you're building
Frames it knows exactly what you're building for this Configuration workflow, including the operating context to verify in the source page.
The autonomy boundary is brutally simple
Frames the autonomy boundary is brutally simple for this Configuration workflow, including the operating context to verify in the source page.
Why "be helpful" doesn't work
Frames why "be helpful" doesn't work for this Configuration workflow, including the operating context to verify in the source page.
Implementation notes
- Use this Configuration flow as a pattern library entry: start from the summary, then inspect the linked source before copying any commands, schedules, or account wiring.
- Primary decision areas: The file that changes everything, Most people train their AI to be useless, It is required to push back, It holds you accountable too, Hermes has a split personality (on purpose). Treat those sections as checkpoints for scope, cost, orchestration, and human review.
- Useful search signals for this flow: 170-line, soul.md, dangerous, single, markdown. These are derived from the public title and summary, not from private runtime data.
Decision table
Use when the summary outcome matches your own workflow and the file that changes everything is relevant to your setup.
Open the source sections for The file that changes everything and Most people train their AI to be useless before wiring credentials or automation.
Keep human approval for merges, spending, external messages, credentials, and unattended execution.
Verification checklist
- Confirm the workflow outcome matches your use case: Why a single 170-line markdown file — not a secret model or magic framework — is what makes a Hermes Agent push back, hold you accountable, and act like an operator instead of a chatbot.
- Open the source section for The file that changes everything before copying commands, prompts, or schedules.
- List every credential, account, model, and external service the flow would touch.
- Define the human approval step before spending money, sending messages, trading, merging, or running unattended.
- Run a small dry run and compare the result with the source sections for The file that changes everything and Most people train their AI to be useless.
Risk notes
Treat every token, OAuth grant, secret manager entry, and API key as production-sensitive. Verify least privilege and revocation before reuse.
Set provider, token, and retry limits before scaling the workflow beyond a single test run.
Check filesystem, shell, browser, repository, and server permissions before granting the agent write access.
What this page covers
- Why a single 170-line markdown file
- not a secret model or magic framework
- is what makes a Hermes Agent push back, hold you accountable
- act like an operator instead of a chatbot.
Source mirror note
This page is generated from the public Hermes Bible index so the clone has the same route coverage and search surface. It stores the public title, category, summary, and source link locally; use the source page for full upstream text and updates.
Open source page