Skip to content

fix(plugins+nous): auto-coerce memory plugins; actionable Nous 401 diagnostic - #14005

Merged
teknium1 merged 2 commits into
mainfrom
hermes/hermes-1a7923fc
Apr 22, 2026
Merged

fix(plugins+nous): auto-coerce memory plugins; actionable Nous 401 diagnostic#14005
teknium1 merged 2 commits into
mainfrom
hermes/hermes-1a7923fc

Conversation

@teknium1

@teknium1 teknium1 commented Apr 22, 2026 •
edited
Loading

Copy link
Copy Markdown
Contributor

Summary

Two small user-reported fixes from Uncle HODL's Discord dump.

1. User memory plugins no longer crash at startup. Bundled memory providers were skipped via skip_names={memory, context_engine}, but plugins at $HERMES_HOME/plugins/<name>/ that call ctx.register_memory_provider(...) were getting dispatched to the general PluginManager (no such method) — AttributeError warning every session.

2. Nous 401 now prints actionable next steps. Previously when Nous returned 401 AND the credential refresh also failed (expired Portal OAuth, out of credits, revoked key), the user saw only a raw APIError with no hint.

Changes

  • hermes_cli/plugins.py _parse_manifest: when kind: isn't explicitly declared, scan __init__.py for register_memory_provider or MemoryProvider (same 8KB text-scan heuristic as plugins/memory/__init__.py:_is_memory_provider_dir) and auto-coerce kind='exclusive'. Routes the plugin to memory discovery. Escape hatch: explicit kind: standalone in manifest is never overridden.
  • run_agent.py Nous 401 branch: on refresh failure, mirror the Anthropic diagnostic block — print the truncated response body, likely causes, and four concrete next steps (hermes login --provider nous, Portal billing link, auth.json path, temporary provider switch).
  • +2 regression tests for the plugin heuristic and its escape hatch.

Validation

Before After
mempalace at startup AttributeError warning every session clean startup, routed to plugins/memory
Explicit kind: standalone with MemoryProvider in comments n/a not coerced
Nous 401 with expired OAuth raw APIError only 5-line diagnostic with hermes login --provider nous as the first step
Plugin tests 75/75 pass 77/77 pass
run_agent auth-related tests — 31/31 pass

Reporter

Uncle HODL [BNKR] on Discord — full dump: https://dpaste.com/4C3Z4AY4F

teknium1 added 2 commits April 22, 2026 05:49
@teknium1
fix(plugins): auto-coerce user-installed memory plugins to kind=exclu…
f0f5288 
…sive

User-installed memory provider plugins at $HERMES_HOME/plugins/<name>/
were being dispatched to the general PluginManager, which has no
register_memory_provider method on PluginContext. Every startup logged:

  Failed to load plugin 'mempalace': 'PluginContext' object has no
  attribute 'register_memory_provider'

Bundled memory providers were already skipped via skip_names={memory,
context_engine} in discover_and_load, but user-installed ones weren't.

Fix: _parse_manifest now scans the plugin's __init__.py source for
'register_memory_provider' or 'MemoryProvider' (same heuristic as
plugins/memory/__init__.py:_is_memory_provider_dir) and auto-coerces
kind to 'exclusive' when the manifest didn't declare one explicitly.
This routes the plugin to plugins/memory discovery instead of the
general loader.

The escape hatch: if a manifest explicitly declares kind: standalone,
the heuristic doesn't override it.

Reported by Uncle HODL on Discord.
@teknium1
fix(nous): actionable CLI message when Nous 401 refresh fails
1e40d3f 
Mirrors the Anthropic 401 diagnostic pattern. When Nous returns 401
and the credential refresh (_try_refresh_nous_client_credentials)
also fails, the user used to see only the raw APIError. Now prints:

  🔐 Nous 401 — Portal authentication failed.
     Response: <truncated body>
     Most likely: Portal OAuth expired, account out of credits, or
                  agent key revoked.
     Troubleshooting:
       • Re-authenticate: hermes login --provider nous
       • Check credits / billing: https://portal.nousresearch.com
       • Verify stored credentials: $HERMES_HOME/auth.json
       • Switch providers temporarily: /model <model> --provider openrouter

Addresses the common 'my hermes model hangs' pattern where the user's
Portal OAuth expired and the CLI gave no hint about the next step.
@teknium1 teknium1 changed the title fix(plugins): auto-coerce user-installed memory plugins to kind=exclusive fix(plugins+nous): auto-coerce memory plugins; actionable Nous 401 diagnostic Apr 22, 2026
@teknium1
teknium1 merged commit 3e652f7 into main Apr 22, 2026
10 of 11 checks passed
@teknium1
teknium1 deleted the hermes/hermes-1a7923fc branch April 22, 2026 12:54
@alt-glitch alt-glitch added type/bug Something isn't working P2 Medium — degraded but workaround exists comp/plugins Plugin system and bundled plugins comp/agent Core agent runtime: loop, agent_init, prompt builder, context-compression, responses endpoint provider/nous Nous Research API (OAuth) labels Apr 22, 2026
@teknium1 teknium1 mentioned this pull request Apr 27, 2026
Closed
aj-nt pushed a commit to aj-nt/hermes-agent that referenced this pull request May 1, 2026
@teknium1
fix(plugins+nous): auto-coerce memory plugins; actionable Nous 401 di…
d01a2fb 
…agnostic (NousResearch#14005)

* fix(plugins): auto-coerce user-installed memory plugins to kind=exclusive

User-installed memory provider plugins at $HERMES_HOME/plugins/<name>/
were being dispatched to the general PluginManager, which has no
register_memory_provider method on PluginContext. Every startup logged:

  Failed to load plugin 'mempalace': 'PluginContext' object has no
  attribute 'register_memory_provider'

Bundled memory providers were already skipped via skip_names={memory,
context_engine} in discover_and_load, but user-installed ones weren't.

Fix: _parse_manifest now scans the plugin's __init__.py source for
'register_memory_provider' or 'MemoryProvider' (same heuristic as
plugins/memory/__init__.py:_is_memory_provider_dir) and auto-coerces
kind to 'exclusive' when the manifest didn't declare one explicitly.
This routes the plugin to plugins/memory discovery instead of the
general loader.

The escape hatch: if a manifest explicitly declares kind: standalone,
the heuristic doesn't override it.

Reported by Uncle HODL on Discord.

* fix(nous): actionable CLI message when Nous 401 refresh fails

Mirrors the Anthropic 401 diagnostic pattern. When Nous returns 401
and the credential refresh (_try_refresh_nous_client_credentials)
also fails, the user used to see only the raw APIError. Now prints:

  🔐 Nous 401 — Portal authentication failed.
     Response: <truncated body>
     Most likely: Portal OAuth expired, account out of credits, or
                  agent key revoked.
     Troubleshooting:
       • Re-authenticate: hermes login --provider nous
       • Check credits / billing: https://portal.nousresearch.com
       • Verify stored credentials: $HERMES_HOME/auth.json
       • Switch providers temporarily: /model <model> --provider openrouter

Addresses the common 'my hermes model hangs' pattern where the user's
Portal OAuth expired and the CLI gave no hint about the next step.
02356abc pushed a commit to 02356abc/hermes-agent that referenced this pull request May 14, 2026
@teknium1
fix(plugins+nous): auto-coerce memory plugins; actionable Nous 401 di…
f237317 
…agnostic (NousResearch#14005)

* fix(plugins): auto-coerce user-installed memory plugins to kind=exclusive

User-installed memory provider plugins at $HERMES_HOME/plugins/<name>/
were being dispatched to the general PluginManager, which has no
register_memory_provider method on PluginContext. Every startup logged:

  Failed to load plugin 'mempalace': 'PluginContext' object has no
  attribute 'register_memory_provider'

Bundled memory providers were already skipped via skip_names={memory,
context_engine} in discover_and_load, but user-installed ones weren't.

Fix: _parse_manifest now scans the plugin's __init__.py source for
'register_memory_provider' or 'MemoryProvider' (same heuristic as
plugins/memory/__init__.py:_is_memory_provider_dir) and auto-coerces
kind to 'exclusive' when the manifest didn't declare one explicitly.
This routes the plugin to plugins/memory discovery instead of the
general loader.

The escape hatch: if a manifest explicitly declares kind: standalone,
the heuristic doesn't override it.

Reported by Uncle HODL on Discord.

* fix(nous): actionable CLI message when Nous 401 refresh fails

Mirrors the Anthropic 401 diagnostic pattern. When Nous returns 401
and the credential refresh (_try_refresh_nous_client_credentials)
also fails, the user used to see only the raw APIError. Now prints:

  🔐 Nous 401 — Portal authentication failed.
     Response: <truncated body>
     Most likely: Portal OAuth expired, account out of credits, or
                  agent key revoked.
     Troubleshooting:
       • Re-authenticate: hermes login --provider nous
       • Check credits / billing: https://portal.nousresearch.com
       • Verify stored credentials: $HERMES_HOME/auth.json
       • Switch providers temporarily: /model <model> --provider openrouter

Addresses the common 'my hermes model hangs' pattern where the user's
Portal OAuth expired and the CLI gave no hint about the next step.
gweeteve pushed a commit to gweeteve/hermes-agent that referenced this pull request Jun 2, 2026
@teknium1
fix(plugins+nous): auto-coerce memory plugins; actionable Nous 401 di…
a0d93a6 
…agnostic (NousResearch#14005)

* fix(plugins): auto-coerce user-installed memory plugins to kind=exclusive

User-installed memory provider plugins at $HERMES_HOME/plugins/<name>/
were being dispatched to the general PluginManager, which has no
register_memory_provider method on PluginContext. Every startup logged:

  Failed to load plugin 'mempalace': 'PluginContext' object has no
  attribute 'register_memory_provider'

Bundled memory providers were already skipped via skip_names={memory,
context_engine} in discover_and_load, but user-installed ones weren't.

Fix: _parse_manifest now scans the plugin's __init__.py source for
'register_memory_provider' or 'MemoryProvider' (same heuristic as
plugins/memory/__init__.py:_is_memory_provider_dir) and auto-coerces
kind to 'exclusive' when the manifest didn't declare one explicitly.
This routes the plugin to plugins/memory discovery instead of the
general loader.

The escape hatch: if a manifest explicitly declares kind: standalone,
the heuristic doesn't override it.

Reported by Uncle HODL on Discord.

* fix(nous): actionable CLI message when Nous 401 refresh fails

Mirrors the Anthropic 401 diagnostic pattern. When Nous returns 401
and the credential refresh (_try_refresh_nous_client_credentials)
also fails, the user used to see only the raw APIError. Now prints:

  🔐 Nous 401 — Portal authentication failed.
     Response: <truncated body>
     Most likely: Portal OAuth expired, account out of credits, or
                  agent key revoked.
     Troubleshooting:
       • Re-authenticate: hermes login --provider nous
       • Check credits / billing: https://portal.nousresearch.com
       • Verify stored credentials: $HERMES_HOME/auth.json
       • Switch providers temporarily: /model <model> --provider openrouter

Addresses the common 'my hermes model hangs' pattern where the user's
Portal OAuth expired and the CLI gave no hint about the next step.
waefrebeorn pushed a commit to waefrebeorn/slermes that referenced this pull request Jul 2, 2026
@teknium1
fix(plugins+nous): auto-coerce memory plugins; actionable Nous 401 di…
c695340 
…agnostic (NousResearch#14005)

* fix(plugins): auto-coerce user-installed memory plugins to kind=exclusive

User-installed memory provider plugins at $HERMES_HOME/plugins/<name>/
were being dispatched to the general PluginManager, which has no
register_memory_provider method on PluginContext. Every startup logged:

  Failed to load plugin 'mempalace': 'PluginContext' object has no
  attribute 'register_memory_provider'

Bundled memory providers were already skipped via skip_names={memory,
context_engine} in discover_and_load, but user-installed ones weren't.

Fix: _parse_manifest now scans the plugin's __init__.py source for
'register_memory_provider' or 'MemoryProvider' (same heuristic as
plugins/memory/__init__.py:_is_memory_provider_dir) and auto-coerces
kind to 'exclusive' when the manifest didn't declare one explicitly.
This routes the plugin to plugins/memory discovery instead of the
general loader.

The escape hatch: if a manifest explicitly declares kind: standalone,
the heuristic doesn't override it.

Reported by Uncle HODL on Discord.

* fix(nous): actionable CLI message when Nous 401 refresh fails

Mirrors the Anthropic 401 diagnostic pattern. When Nous returns 401
and the credential refresh (_try_refresh_nous_client_credentials)
also fails, the user used to see only the raw APIError. Now prints:

  🔐 Nous 401 — Portal authentication failed.
     Response: <truncated body>
     Most likely: Portal OAuth expired, account out of credits, or
                  agent key revoked.
     Troubleshooting:
       • Re-authenticate: hermes login --provider nous
       • Check credits / billing: https://portal.nousresearch.com
       • Verify stored credentials: $HERMES_HOME/auth.json
       • Switch providers temporarily: /model <model> --provider openrouter

Addresses the common 'my hermes model hangs' pattern where the user's
Portal OAuth expired and the CLI gave no hint about the next step.
smasoftware pushed a commit to smasoftware/hermes-agent that referenced this pull request Jul 18, 2026
@teknium1 @smasoftware
fix(plugins+nous): auto-coerce memory plugins; actionable Nous 401 di…
545d91f 
…agnostic (NousResearch#14005)

* fix(plugins): auto-coerce user-installed memory plugins to kind=exclusive

User-installed memory provider plugins at $HERMES_HOME/plugins/<name>/
were being dispatched to the general PluginManager, which has no
register_memory_provider method on PluginContext. Every startup logged:

  Failed to load plugin 'mempalace': 'PluginContext' object has no
  attribute 'register_memory_provider'

Bundled memory providers were already skipped via skip_names={memory,
context_engine} in discover_and_load, but user-installed ones weren't.

Fix: _parse_manifest now scans the plugin's __init__.py source for
'register_memory_provider' or 'MemoryProvider' (same heuristic as
plugins/memory/__init__.py:_is_memory_provider_dir) and auto-coerces
kind to 'exclusive' when the manifest didn't declare one explicitly.
This routes the plugin to plugins/memory discovery instead of the
general loader.

The escape hatch: if a manifest explicitly declares kind: standalone,
the heuristic doesn't override it.

Reported by Uncle HODL on Discord.

* fix(nous): actionable CLI message when Nous 401 refresh fails

Mirrors the Anthropic 401 diagnostic pattern. When Nous returns 401
and the credential refresh (_try_refresh_nous_client_credentials)
also fails, the user used to see only the raw APIError. Now prints:

  🔐 Nous 401 — Portal authentication failed.
     Response: <truncated body>
     Most likely: Portal OAuth expired, account out of credits, or
                  agent key revoked.
     Troubleshooting:
       • Re-authenticate: hermes login --provider nous
       • Check credits / billing: https://portal.nousresearch.com
       • Verify stored credentials: $HERMES_HOME/auth.json
       • Switch providers temporarily: /model <model> --provider openrouter

Addresses the common 'my hermes model hangs' pattern where the user's
Portal OAuth expired and the CLI gave no hint about the next step.
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Reviewers

No reviews

Assignees

No one assigned

Labels

comp/agent Core agent runtime: loop, agent_init, prompt builder, context-compression, responses endpoint comp/plugins Plugin system and bundled plugins P2 Medium — degraded but workaround exists provider/nous Nous Research API (OAuth) type/bug Something isn't working

Projects

None yet

Milestone

No milestone

Development

Successfully merging this pull request may close these issues.

2 participants

@teknium1 @alt-glitch